diff --git a/.gitignore b/.gitignore
index 7efef52..ffb00ec 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,3 +1,16 @@
 adni_storage/
+abide_storage/
+cobre_storage/
+corr_storage/
+dlbs_storage/
+fcon1000_storage/
+ixi_storage/
+oasis1_storage/
+sald_storage/
 ants/
-model_dumps/
\ No newline at end of file
+model_dumps/
+pretrained/
+medvit/
+processed_slices/
+vit_trained_bak/
+camcan_storage/
diff --git a/3dres.py b/3dres.py
new file mode 100644
index 0000000..e69de29
diff --git a/Papers/s41598-024-59578-3.pdf b/Papers/s41598-024-59578-3.pdf
new file mode 100644
index 0000000..75f1274
Binary files /dev/null and b/Papers/s41598-024-59578-3.pdf differ
diff --git a/__pycache__/classify_fnn.cpython-312.pyc b/__pycache__/classify_fnn.cpython-312.pyc
new file mode 100644
index 0000000..394f925
Binary files /dev/null and b/__pycache__/classify_fnn.cpython-312.pyc differ
diff --git a/__pycache__/classify_fnn_bag.cpython-312.pyc b/__pycache__/classify_fnn_bag.cpython-312.pyc
new file mode 100644
index 0000000..26e87d0
Binary files /dev/null and b/__pycache__/classify_fnn_bag.cpython-312.pyc differ
diff --git a/__pycache__/cnn_gn.cpython-312.pyc b/__pycache__/cnn_gn.cpython-312.pyc
new file mode 100644
index 0000000..0edbe6a
Binary files /dev/null and b/__pycache__/cnn_gn.cpython-312.pyc differ
diff --git a/__pycache__/cnn_mx.cpython-312.pyc b/__pycache__/cnn_mx.cpython-312.pyc
new file mode 100644
index 0000000..ad1995a
Binary files /dev/null and b/__pycache__/cnn_mx.cpython-312.pyc differ
diff --git a/__pycache__/cnn_mx_att.cpython-312.pyc b/__pycache__/cnn_mx_att.cpython-312.pyc
new file mode 100644
index 0000000..bda57d5
Binary files /dev/null and b/__pycache__/cnn_mx_att.cpython-312.pyc differ
diff --git a/__pycache__/cnn_mx_att_bl.cpython-312.pyc b/__pycache__/cnn_mx_att_bl.cpython-312.pyc
new file mode 100644
index 0000000..668693e
Binary files /dev/null and b/__pycache__/cnn_mx_att_bl.cpython-312.pyc differ
diff --git a/__pycache__/cnn_mx_att_gate2.cpython-312.pyc b/__pycache__/cnn_mx_att_gate2.cpython-312.pyc
new file mode 100644
index 0000000..c35082d
Binary files /dev/null and b/__pycache__/cnn_mx_att_gate2.cpython-312.pyc differ
diff --git a/__pycache__/cnn_mx_att_res.cpython-312.pyc b/__pycache__/cnn_mx_att_res.cpython-312.pyc
new file mode 100644
index 0000000..2dbbbf9
Binary files /dev/null and b/__pycache__/cnn_mx_att_res.cpython-312.pyc differ
diff --git a/__pycache__/cnn_mx_cbam.cpython-312.pyc b/__pycache__/cnn_mx_cbam.cpython-312.pyc
new file mode 100644
index 0000000..5bf30af
Binary files /dev/null and b/__pycache__/cnn_mx_cbam.cpython-312.pyc differ
diff --git a/__pycache__/cnn_mx_elu.cpython-312.pyc b/__pycache__/cnn_mx_elu.cpython-312.pyc
new file mode 100644
index 0000000..eb3e205
Binary files /dev/null and b/__pycache__/cnn_mx_elu.cpython-312.pyc differ
diff --git a/__pycache__/cnn_mx_lrelu.cpython-312.pyc b/__pycache__/cnn_mx_lrelu.cpython-312.pyc
new file mode 100644
index 0000000..7b34fbe
Binary files /dev/null and b/__pycache__/cnn_mx_lrelu.cpython-312.pyc differ
diff --git a/__pycache__/cnn_mx_multi_att.cpython-312.pyc b/__pycache__/cnn_mx_multi_att.cpython-312.pyc
new file mode 100644
index 0000000..d9a79c2
Binary files /dev/null and b/__pycache__/cnn_mx_multi_att.cpython-312.pyc differ
diff --git a/__pycache__/cnn_mx_res.cpython-312.pyc b/__pycache__/cnn_mx_res.cpython-312.pyc
new file mode 100644
index 0000000..5e85ff3
Binary files /dev/null and b/__pycache__/cnn_mx_res.cpython-312.pyc differ
diff --git a/__pycache__/cnn_mx_spd.cpython-312.pyc b/__pycache__/cnn_mx_spd.cpython-312.pyc
new file mode 100644
index 0000000..d2547f4
Binary files /dev/null and b/__pycache__/cnn_mx_spd.cpython-312.pyc differ
diff --git a/__pycache__/cnn_mx_sw.cpython-312.pyc b/__pycache__/cnn_mx_sw.cpython-312.pyc
new file mode 100644
index 0000000..61a9571
Binary files /dev/null and b/__pycache__/cnn_mx_sw.cpython-312.pyc differ
diff --git a/__pycache__/cnn_mx_vl.cpython-312.pyc b/__pycache__/cnn_mx_vl.cpython-312.pyc
new file mode 100644
index 0000000..2a5ad8e
Binary files /dev/null and b/__pycache__/cnn_mx_vl.cpython-312.pyc differ
diff --git a/__pycache__/dataset_cls.cpython-312.pyc b/__pycache__/dataset_cls.cpython-312.pyc
new file mode 100644
index 0000000..b9e8205
Binary files /dev/null and b/__pycache__/dataset_cls.cpython-312.pyc differ
diff --git a/adni_preprocess.sh b/adni_preprocess.sh
index 3071cf7..224ecb0 100644
--- a/adni_preprocess.sh
+++ b/adni_preprocess.sh
@@ -4,11 +4,12 @@
 count=0
 
 # Define source and target directories
-source_folder="adni_storage/ADNI_nii_gz_stripped"
-target_folder="adni_storage/ADNI_nii_gz_bias_corrected"
+source_folder="camcan_storage/CamCAN_nii_gz_stripped"
+target_folder="camcan_storage/CamCAN_nii_gz_bias_corrected"
 
 # Path to N4BiasFieldCorrection binary
-n4bias_path="ants/ants-2.5.4/bin/N4BiasFieldCorrection"
+# n4bias_path="ants/ants-2.5.4/bin/N4BiasFieldCorrection"
+n4bias_path="~/miniconda3/envs/db3/bin/N4BiasFieldCorrection"
 
 # Loop through all .nii.gz files in the source folder
 for input_file in "${source_folder}"/*.stripped.nii.gz; do
diff --git a/adni_storage b/adni_storage
deleted file mode 120000
index f714944..0000000
--- a/adni_storage
+++ /dev/null
@@ -1 +0,0 @@
-/NTFS-HDD/Storage/adnii_storage
\ No newline at end of file
diff --git a/age_distribution.png b/age_distribution.png
new file mode 100644
index 0000000..e44a029
Binary files /dev/null and b/age_distribution.png differ
diff --git a/calculate_agegap.py b/calculate_agegap.py
index 623af3f..648e456 100644
--- a/calculate_agegap.py
+++ b/calculate_agegap.py
@@ -38,7 +38,7 @@ def plot_with_metrics(data, x_col, y_col, hue_col, title, x_lim):
     plt.show()
 
 # For training set
-dfres_train = pd.read_csv("predicted_ages_train.csv", sep=",", index_col=0).reset_index()
+dfres_train = pd.read_csv("model_dumps/cnn_mx_elu_predicted_ages_train.csv", sep=",", index_col=0).reset_index()
 dfres_train = calculate_lowess_yhat_and_agegap(dfres_train)
 
 # Keep only the row with the smallest Age for each SubjectID
@@ -50,7 +50,7 @@ def plot_with_metrics(data, x_col, y_col, hue_col, title, x_lim):
                   title="Age gap predictions (Train Set)", x_lim=(40, 100))
 
 # For validation set
-dfres_val = pd.read_csv("predicted_ages_val.csv", sep=",", index_col=0).reset_index()
+dfres_val = pd.read_csv("model_dumps/cnn_mx_elu_predicted_ages_val.csv", sep=",", index_col=0).reset_index()
 dfres_val = calculate_lowess_yhat_and_agegap(dfres_val)
 
 # Keep only the row with the smallest Age for each SubjectID
diff --git a/calculate_agegap_da.py b/calculate_agegap_da.py
new file mode 100644
index 0000000..2170b24
--- /dev/null
+++ b/calculate_agegap_da.py
@@ -0,0 +1,186 @@
+import pandas as pd
+import numpy as np
+import statsmodels.api as sm
+import seaborn as sns
+from scipy.stats import norm
+from sklearn.metrics import mean_squared_error, mean_absolute_error, r2_score
+from scipy.interpolate import make_interp_spline, interp1d
+import matplotlib.pyplot as plt
+from sklearn.model_selection import train_test_split
+from sklearn.svm import SVC  # Import SVM
+from sklearn.preprocessing import LabelEncoder
+from sklearn.metrics import classification_report, confusion_matrix
+from sklearn.mixture import GaussianMixture
+from sklearn.metrics import classification_report, confusion_matrix, accuracy_score
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, TensorDataset
+
+def calculate_lowess_yhat_and_agegap(dfres):
+    dfres_agegap = dfres.copy()
+    # calculate agegap using lowess of predicted vs chronological age from training cohort
+    lowess = sm.nonparametric.lowess
+    lowess_fit = lowess(dfres_agegap.Predicted_Age.to_numpy(), dfres_agegap.Age.to_numpy(), frac=0.8, it=3)
+    lowess_fit_int = interp1d(lowess_fit[:,0], lowess_fit[:,1], bounds_error=False, kind='linear', fill_value=(0, 150)) 
+    y_lowess = lowess_fit_int(dfres_agegap.Age)
+    dfres_agegap["yhat_lowess"] = y_lowess
+    # dfres_agegap["yhat_lowess"] = age_prediction_lowess(np.array(dfres_agegap.Age))
+    if len(dfres_agegap.loc[dfres_agegap.yhat_lowess.isna()]) > 0:
+        print("Could not predict lowess yhat in " + str(len(dfres_agegap.loc[dfres_agegap.yhat_lowess.isna()])) + " samples")
+        dfres_agegap = dfres_agegap.dropna(subset="yhat_lowess")
+    dfres_agegap["AgeGap"] = dfres_agegap["Predicted_Age"] - dfres_agegap["yhat_lowess"]
+    dfres_agegap["AgeGap"] = dfres_agegap["AgeGap"].abs()
+    return dfres_agegap
+
+# Function to calculate MAE and R², and annotate the plot
+def plot_with_metrics(data, x_col, y_col, hue_col, title, x_lim):
+    # Calculate MAE and R²
+    mae = mean_absolute_error(data[x_col], data[y_col])
+    r2 = r2_score(data[x_col], data[y_col])
+    
+    # Create scatterplot
+    sns.scatterplot(data=data, x=x_col, y=y_col, hue=hue_col, palette='coolwarm', hue_norm=(-12, 12))
+    plt.xlim(*x_lim)
+    plt.title(f"{title}\nMAE: {mae:.2f}, R²: {r2:.2f}")
+    plt.xlabel(x_col)
+    plt.ylabel(y_col)
+    plt.show()
+
+# For training set
+dfres_train = pd.read_csv("model_dumps/cnn_mx_elu_predicted_ages_train.csv", sep=",", index_col=0).reset_index()
+dfres_train = calculate_lowess_yhat_and_agegap(dfres_train)
+
+# Keep only the row with the smallest Age for each SubjectID
+dfres_train = dfres_train.loc[dfres_train.groupby('SubjectID')['Age'].idxmin()]
+dfres_train = dfres_train.reset_index(drop=True)
+
+# For validation set
+dfres_val = pd.read_csv("model_dumps/cnn_mx_elu_predicted_ages_val.csv", sep=",", index_col=0).reset_index()
+dfres_val = calculate_lowess_yhat_and_agegap(dfres_val)
+
+# Keep only the row with the smallest Age for each SubjectID
+dfres_val = dfres_val.loc[dfres_val.groupby('SubjectID')['Age'].idxmin()]
+dfres_val = dfres_val.reset_index(drop=True)
+
+
+# Step 1: Encode categorical variables (for 'Sex' column)
+dfres_train['Sex'] = dfres_train['Sex'].map({'M': 0, 'F': 1})
+dfres_val['Sex'] = dfres_val['Sex'].map({'M': 0, 'F': 1})
+
+# Step 2: Convert the 'Group' column to binary (AD vs not AD)
+dfres_train['Group_binary'] = dfres_train['Group'].apply(lambda x: 1 if x == 'AD' else 0)
+dfres_val['Group_binary'] = dfres_val['Group'].apply(lambda x: 1 if x == 'AD' else 0)
+
+# Step 3: Initialize the LabelEncoder for the binary target 'Group_binary' column
+y_train = dfres_train['Group_binary']
+y_val = dfres_val['Group_binary']
+
+print(f"Binary labels for training set: {y_train.unique()}")  # To verify the binary classification
+
+# Step 4: Drop the original 'Group' column and prepare features for training
+X_train = dfres_train[['AgeGap']]
+X_val = dfres_val[['AgeGap']]
+
+print(f"Features for training set:\n{X_train.head()}")
+
+# Step 1: Prepare the data
+X_train = torch.tensor(dfres_train[['AgeGap']].values, dtype=torch.float32)  # Feature
+y_train = torch.tensor(dfres_train['Group_binary'].values, dtype=torch.float32)  # Target (binary)
+
+X_val = torch.tensor(dfres_val[['AgeGap']].values, dtype=torch.float32)  # Feature
+y_val = torch.tensor(dfres_val['Group_binary'].values, dtype=torch.float32)  # Target (binary)
+
+# Step 2: Create Dataloader
+train_dataset = TensorDataset(X_train, y_train)
+val_dataset = TensorDataset(X_val, y_val)
+
+train_loader = DataLoader(train_dataset, batch_size=32, shuffle=True)
+val_loader = DataLoader(val_dataset, batch_size=32, shuffle=False)
+
+# Step 3: Define the Model with Attention Mechanism
+class AttentionDNN(nn.Module):
+    def __init__(self):
+        super(AttentionDNN, self).__init__()
+        self.dense1 = nn.Linear(1, 64)
+        self.dropout1 = nn.Dropout(0.2)
+        
+        # Attention mechanism (simple scaled dot-product attention)
+        self.attention = nn.MultiheadAttention(embed_dim=64, num_heads=1, batch_first=True)
+        
+        self.dense2 = nn.Linear(128, 32)  # 64 + 64 (Concatenated)
+        self.dropout2 = nn.Dropout(0.2)
+        self.output = nn.Linear(32, 1)  # Binary output
+
+    def forward(self, x):
+        # First dense layer
+        x = torch.relu(self.dense1(x))
+        x = self.dropout1(x).unsqueeze(1)  # Add extra dimension for attention (batch_size, seq_len, feature)
+        
+        # Attention mechanism (requires 3 inputs: queries, keys, values)
+        attn_output, _ = self.attention(x, x, x)
+        
+        # Concatenate attention output with original features
+        x = torch.cat((x, attn_output), dim=-1)
+        
+        # Second dense layer
+        x = torch.relu(self.dense2(x))
+        x = self.dropout2(x)
+        
+        # Output layer (sigmoid for binary classification)
+        x = torch.sigmoid(self.output(x.squeeze(1)))  # Remove extra dimension
+        
+        return x
+
+# Step 4: Initialize the model, loss function, and optimizer
+model = AttentionDNN()
+criterion = nn.BCELoss()  # Binary Cross Entropy loss for binary classification
+optimizer = optim.Adam(model.parameters(), lr=0.001)
+
+# Step 5: Train the model
+num_epochs = 100
+for epoch in range(num_epochs):
+    model.train()
+    running_loss = 0.0
+    for inputs, labels in train_loader:
+        optimizer.zero_grad()
+        
+        outputs = model(inputs)
+        loss = criterion(outputs, labels.view(-1, 1))  # Reshaping labels for BCELoss
+        loss.backward()
+        optimizer.step()
+        
+        running_loss += loss.item()
+    
+    print(f"Epoch [{epoch+1}/{num_epochs}], Loss: {running_loss/len(train_loader)}")
+
+# Step 6: Evaluate the model
+model.eval()
+y_pred_prob = []
+y_true = []
+
+with torch.no_grad():
+    for inputs, labels in val_loader:
+        outputs = model(inputs)
+        y_pred_prob.append(outputs)
+        y_true.append(labels)
+
+y_pred_prob = torch.cat(y_pred_prob)
+y_true = torch.cat(y_true)
+
+# Convert to binary labels (0 or 1)
+y_pred = (y_pred_prob > 0.5).float()
+
+# Classification Report
+class_names = ['Not AD', 'AD']
+print("Classification Report:")
+print(classification_report(y_true, y_pred, target_names=class_names))
+
+# Confusion Matrix
+conf_matrix = confusion_matrix(y_true, y_pred)
+plt.figure(figsize=(8, 6))
+sns.heatmap(conf_matrix, annot=True, fmt='d', cmap='Blues', xticklabels=class_names, yticklabels=class_names)
+plt.title('Confusion Matrix')
+plt.xlabel('Predicted Labels')
+plt.ylabel('True Labels')
+plt.show()
diff --git a/calculate_agegap_gmm.py b/calculate_agegap_gmm.py
new file mode 100644
index 0000000..89287a2
--- /dev/null
+++ b/calculate_agegap_gmm.py
@@ -0,0 +1,114 @@
+import pandas as pd
+import numpy as np
+import statsmodels.api as sm
+import seaborn as sns
+from scipy.stats import norm
+from sklearn.metrics import mean_squared_error, mean_absolute_error, r2_score
+from scipy.interpolate import make_interp_spline, interp1d
+import matplotlib.pyplot as plt
+from sklearn.model_selection import train_test_split
+from sklearn.svm import SVC  # Import SVM
+from sklearn.preprocessing import LabelEncoder
+from sklearn.metrics import classification_report, confusion_matrix
+from sklearn.mixture import GaussianMixture
+from sklearn.metrics import classification_report, confusion_matrix, accuracy_score
+
+
+
+def calculate_lowess_yhat_and_agegap(dfres):
+    dfres_agegap = dfres.copy()
+    # calculate agegap using lowess of predicted vs chronological age from training cohort
+    lowess = sm.nonparametric.lowess
+    lowess_fit = lowess(dfres_agegap.Predicted_Age.to_numpy(), dfres_agegap.Age.to_numpy(), frac=0.8, it=3)
+    lowess_fit_int = interp1d(lowess_fit[:,0], lowess_fit[:,1], bounds_error=False, kind='linear', fill_value=(0, 150)) 
+    y_lowess = lowess_fit_int(dfres_agegap.Age)
+    dfres_agegap["yhat_lowess"] = y_lowess
+    # dfres_agegap["yhat_lowess"] = age_prediction_lowess(np.array(dfres_agegap.Age))
+    if len(dfres_agegap.loc[dfres_agegap.yhat_lowess.isna()]) > 0:
+        print("Could not predict lowess yhat in " + str(len(dfres_agegap.loc[dfres_agegap.yhat_lowess.isna()])) + " samples")
+        dfres_agegap = dfres_agegap.dropna(subset="yhat_lowess")
+    dfres_agegap["AgeGap"] = dfres_agegap["Predicted_Age"] - dfres_agegap["yhat_lowess"]
+    dfres_agegap["AgeGap"] = dfres_agegap["AgeGap"].abs()
+    return dfres_agegap
+
+# Function to calculate MAE and R², and annotate the plot
+def plot_with_metrics(data, x_col, y_col, hue_col, title, x_lim):
+    # Calculate MAE and R²
+    mae = mean_absolute_error(data[x_col], data[y_col])
+    r2 = r2_score(data[x_col], data[y_col])
+    
+    # Create scatterplot
+    sns.scatterplot(data=data, x=x_col, y=y_col, hue=hue_col, palette='coolwarm', hue_norm=(-12, 12))
+    plt.xlim(*x_lim)
+    plt.title(f"{title}\nMAE: {mae:.2f}, R²: {r2:.2f}")
+    plt.xlabel(x_col)
+    plt.ylabel(y_col)
+    plt.show()
+
+# For training set
+dfres_train = pd.read_csv("model_dumps/cnn_mx_elu_predicted_ages_train.csv", sep=",", index_col=0).reset_index()
+dfres_train = calculate_lowess_yhat_and_agegap(dfres_train)
+
+# Keep only the row with the smallest Age for each SubjectID
+dfres_train = dfres_train.loc[dfres_train.groupby('SubjectID')['Age'].idxmin()]
+dfres_train = dfres_train.reset_index(drop=True)
+
+# For validation set
+dfres_val = pd.read_csv("model_dumps/cnn_mx_elu_predicted_ages_val.csv", sep=",", index_col=0).reset_index()
+dfres_val = calculate_lowess_yhat_and_agegap(dfres_val)
+
+# Keep only the row with the smallest Age for each SubjectID
+dfres_val = dfres_val.loc[dfres_val.groupby('SubjectID')['Age'].idxmin()]
+dfres_val = dfres_val.reset_index(drop=True)
+
+
+# Step 1: Encode categorical variables (for 'Sex' column)
+dfres_train['Sex'] = dfres_train['Sex'].map({'M': 0, 'F': 1})
+dfres_val['Sex'] = dfres_val['Sex'].map({'M': 0, 'F': 1})
+
+# Step 2: Convert the 'Group' column to binary (AD vs not AD)
+dfres_train['Group_binary'] = dfres_train['Group'].apply(lambda x: 1 if x == 'AD' else 0)
+dfres_val['Group_binary'] = dfres_val['Group'].apply(lambda x: 1 if x == 'AD' else 0)
+
+# Step 3: Initialize the LabelEncoder for the binary target 'Group_binary' column
+y_train = dfres_train['Group_binary']
+y_val = dfres_val['Group_binary']
+
+print(f"Binary labels for training set: {y_train.unique()}")  # To verify the binary classification
+
+# Step 4: Drop the original 'Group' column and prepare features for training
+X_train = dfres_train[['AgeGap']]
+X_val = dfres_val[['AgeGap']]
+
+print(f"Features for training set:\n{X_train.head()}")
+
+# Step 5: Train the Gaussian Mixture Model (GMM)
+gmm = GaussianMixture(n_components=2, covariance_type='full', random_state=42)
+gmm.fit(X_train)
+
+# Step 6: Predict probabilities and classify
+# GMM provides probabilities for each component (class). Choose the class with the highest probability.
+y_pred_prob = gmm.predict_proba(X_val)
+y_pred = np.argmax(y_pred_prob, axis=1)  # Get the class with the highest probability
+
+# Get the class labels (binary: 0 = not AD, 1 = AD)
+class_names = ['Not AD', 'AD']
+
+# Step 7: Evaluate the GMM model
+print("Classification Report:")
+print(classification_report(y_val, y_pred, target_names=class_names))
+
+print("Confusion Matrix:")
+conf_matrix = confusion_matrix(y_val, y_pred)
+
+# Create a heatmap with seaborn
+plt.figure(figsize=(8, 6))
+sns.heatmap(conf_matrix, annot=True, fmt='d', cmap='Blues', xticklabels=class_names, yticklabels=class_names)
+plt.title('Confusion Matrix')
+plt.xlabel('Predicted Labels')
+plt.ylabel('True Labels')
+plt.show()
+
+# Step 8: (Optional) Evaluate overall accuracy
+accuracy = accuracy_score(y_val, y_pred)
+print(f"Accuracy: {accuracy:.2f}")
\ No newline at end of file
diff --git a/classify_fnn.py b/classify_fnn.py
new file mode 100644
index 0000000..f7a8e3e
--- /dev/null
+++ b/classify_fnn.py
@@ -0,0 +1,80 @@
+
+
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, Dataset
+
+class AgePredictionCNNMulticlass(nn.Module):
+    def __init__(self, input_shape, num_classes):
+        super(AgePredictionCNNMulticlass, self).__init__()
+
+        # Define convolutional and pooling layers
+        self.conv1 = nn.Conv2d(1, 1, kernel_size=(10, 60), stride=1)
+        self.pool1 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+        self.conv2 = nn.Conv2d(1, 1, kernel_size=(5, 15), stride=1)
+        self.pool2 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+        self.conv3 = nn.Conv2d(1, 1, kernel_size=(2, 6), stride=1)
+        self.pool3 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+        self.flatten = nn.Flatten()
+
+        # Fully connected layers
+        self.fc1 = None  # Placeholder to be initialized dynamically
+        self.fc1_bn = None  # Placeholder for batch normalization after fc1
+        self.fc2 = nn.Linear(512, 128)
+        self.fc2_bn = nn.LayerNorm(128)
+        self.dropout = nn.Dropout(p=0.1)  # Dropout with 10% probability
+        self.fc3 = nn.Linear(129, num_classes)  # Output for multiple classes
+        
+        self.lrelu = nn.LeakyReLU()
+        self.initialize_fc1(input_shape)
+
+    def initialize_fc1(self, input_shape):
+        # Create a sample input to pass through the convolutional layers
+        sample_input = torch.zeros(1, *input_shape)
+        x = self.conv1(sample_input)
+        x = self.lrelu(x)
+        x = self.pool1(x)  # Apply pooling
+        x = self.conv2(x)
+        x = self.lrelu(x)
+        x = self.pool2(x)  # Apply pooling
+        x = self.conv3(x)
+        x = self.lrelu(x)
+        x = self.pool3(x)  # Apply pooling
+        flattened_size = x.numel()  # Total number of elements after flattening
+        self.fc1 = nn.Linear(flattened_size, 512)
+        self.fc1_bn = nn.LayerNorm(512)  # Initialize batch normalization for fc1
+
+    def forward(self, x, sex):
+        x = self.conv1(x)
+        x = self.lrelu(x)
+        x = self.pool1(x)  # Apply pooling
+        x = self.conv2(x)
+        x = self.lrelu(x)
+        x = self.pool2(x)  # Apply pooling
+        x = self.conv3(x)
+        x = self.lrelu(x)
+        x = self.pool3(x)  # Apply pooling
+        x = self.flatten(x)
+
+        if self.fc1 is None:
+            raise ValueError("fc1 layer has not been initialized. Call `initialize_fc1` with the input shape.")
+
+        x = self.fc1(x)
+        x = self.fc1_bn(x)  # Apply batch normalization
+        x = self.lrelu(x)
+        x = self.dropout(x)  # Apply dropout
+
+        x = self.fc2(x)
+        x = self.fc2_bn(x)  # Apply batch normalization
+        x = self.lrelu(x)
+        x = self.dropout(x)  # Apply dropout
+
+        # Concatenate `Sex` input
+        x = torch.cat((x, sex.unsqueeze(1)), dim=1)
+        x = self.fc3(x)
+
+        return x
\ No newline at end of file
diff --git a/classify_fnn_bag.py b/classify_fnn_bag.py
new file mode 100644
index 0000000..2012c2e
--- /dev/null
+++ b/classify_fnn_bag.py
@@ -0,0 +1,82 @@
+
+
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, Dataset
+
+class AgePredictionCNNMulticlass(nn.Module):
+    def __init__(self, input_shape, num_classes):
+        super(AgePredictionCNNMulticlass, self).__init__()
+
+        # Define convolutional and pooling layers
+        self.conv1 = nn.Conv2d(1, 1, kernel_size=(10, 60), stride=1)
+        self.pool1 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+        self.conv2 = nn.Conv2d(1, 1, kernel_size=(5, 15), stride=1)
+        self.pool2 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+        self.conv3 = nn.Conv2d(1, 1, kernel_size=(2, 6), stride=1)
+        self.pool3 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+        self.flatten = nn.Flatten()
+
+        # Fully connected layers
+        self.fc1 = None  # Placeholder to be initialized dynamically
+        self.fc1_bn = None  # Placeholder for batch normalization after fc1
+        self.fc2 = nn.Linear(512, 128)
+        self.fc2_bn = nn.LayerNorm(128)
+        self.dropout = nn.Dropout(p=0.1)  # Dropout with 10% probability
+        self.fc3 = nn.Linear(130, num_classes)  # Output for multiple classes
+        
+        self.lrelu = nn.LeakyReLU()
+        self.initialize_fc1(input_shape)
+
+    def initialize_fc1(self, input_shape):
+        # Create a sample input to pass through the convolutional layers
+        sample_input = torch.zeros(1, *input_shape)
+        x = self.conv1(sample_input)
+        x = self.lrelu(x)
+        x = self.pool1(x)  # Apply pooling
+        x = self.conv2(x)
+        x = self.lrelu(x)
+        x = self.pool2(x)  # Apply pooling
+        x = self.conv3(x)
+        x = self.lrelu(x)
+        x = self.pool3(x)  # Apply pooling
+        flattened_size = x.numel()  # Total number of elements after flattening
+        self.fc1 = nn.Linear(flattened_size, 512)
+        self.fc1_bn = nn.LayerNorm(512)  # Initialize batch normalization for fc1
+
+    def forward(self, x, sex, age_gap):
+        x = self.conv1(x)
+        x = self.lrelu(x)
+        x = self.pool1(x)  # Apply pooling
+        x = self.conv2(x)
+        x = self.lrelu(x)
+        x = self.pool2(x)  # Apply pooling
+        x = self.conv3(x)
+        x = self.lrelu(x)
+        x = self.pool3(x)  # Apply pooling
+        x = self.flatten(x)
+
+        if self.fc1 is None:
+            raise ValueError("fc1 layer has not been initialized. Call `initialize_fc1` with the input shape.")
+
+        x = self.fc1(x)
+        x = self.fc1_bn(x)  # Apply batch normalization
+        x = self.lrelu(x)
+        x = self.dropout(x)  # Apply dropout
+
+        x = self.fc2(x)
+        x = self.fc2_bn(x)  # Apply batch normalization
+        x = self.lrelu(x)
+        x = self.dropout(x)  # Apply dropout
+
+        # Concatenate `Sex` input
+        x = torch.cat((x, sex.unsqueeze(1)), dim=1)
+        # Concatenate `Age Gap` input
+        x = torch.cat((x, age_gap.unsqueeze(1)), dim=1)
+        x = self.fc3(x)
+
+        return x
\ No newline at end of file
diff --git a/classify_main.py b/classify_main.py
new file mode 100644
index 0000000..aff0089
--- /dev/null
+++ b/classify_main.py
@@ -0,0 +1,312 @@
+import pandas as pd
+import numpy as np
+import matplotlib.pyplot as plt
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler
+from sklearn.metrics import confusion_matrix
+from sklearn.metrics import accuracy_score
+import torch
+import torch.nn as nn
+import torch.nn.init as init
+import torch.optim as optim
+from tqdm import tqdm
+from torchvision import transforms
+from transformers import ViTFeatureExtractor, ViTModel
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler
+import pickle
+import matplotlib.pyplot as plt
+import sys
+import os
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, Dataset
+import seaborn as sns
+
+def set_random_seed(seed=69420):
+    np.random.seed(seed)
+    torch.manual_seed(seed)
+    torch.cuda.manual_seed(seed)
+    torch.cuda.manual_seed_all(seed)  # For multi-GPU setups
+    torch.backends.cudnn.deterministic = True
+    torch.backends.cudnn.benchmark = False
+
+set_random_seed(69420)
+np.random.seed(69420)
+
+# Load the data
+csv_path = "adni_storage/adni_brainrotnet_metadata.csv"
+df = pd.read_csv(csv_path)
+df['filepath'] = df.apply(
+    lambda row: f"adni_storage/ADNI_nii_gz_bias_corrected/I{row['ImageID']}_{row['SubjectID']}.stripped.N4.nii.gz",
+    axis=1
+)
+
+# convert the Group column to numeric
+df['Group'] = df['Group'].map({'CN': 0, 'AD': 1, 'MCI': 2})
+sex_encoded = df['Sex'].apply(lambda x: 0 if x == 'M' else 1).tolist()
+
+features_list = []
+labels_list = []
+
+for _, row in tqdm(df.iterrows(), total=len(df), desc="Processing images"):
+    filepath = row['filepath']
+    image_title = f"{row['ImageID']}_{row['SubjectID']}"
+
+    # Check if the feature file already exists
+    feature_file_path = f"adni_storage/ADNI_features/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Group'])  # Add the corresponding age label
+
+# Convert the lists to numpy arrays
+features = np.array(features_list)
+labels = np.array(labels_list)
+
+
+# Custom Dataset
+class ADNIDataset(Dataset):
+    def __init__(self, features_list, sex_list, labels_list):
+        self.features = features_list
+        self.sex = sex_list
+        self.labels = labels_list
+   
+
+    def __len__(self):
+        return len(self.features)
+
+    def __getitem__(self, idx):
+        return (
+            torch.tensor(self.features[idx], dtype=torch.float32),
+            torch.tensor(self.sex[idx], dtype=torch.float32),
+            torch.tensor(self.labels[idx], dtype=torch.long),
+        )
+    
+# Create Dataset and DataLoader
+dataset = ADNIDataset(features_list, sex_encoded, labels_list)
+train_size = int(0.8 * len(dataset))
+val_size = len(dataset) - train_size
+train_dataset, val_dataset = torch.utils.data.random_split(dataset, [train_size, val_size])
+
+train_loader = DataLoader(train_dataset, batch_size=1, shuffle=False)
+val_loader = DataLoader(val_dataset, batch_size=1, shuffle=False)
+
+
+import importlib
+
+module_name = sys.argv[1]  # Example: "my_model"
+class_name = "AgePredictionCNNMulticlass"  # The class you want to import
+
+try:
+    # Dynamically import the module
+    module = importlib.import_module(module_name)
+    
+    # Dynamically get the class
+    AgePredictionCNNMulticlass = getattr(module, class_name)
+    
+    print(f"Successfully imported {class_name} from {module_name}.")
+
+except ImportError:
+    print(f"Module {module_name} could not be imported.")
+except AttributeError:
+    print(f"{class_name} does not exist in {module_name}.")
+
+model = AgePredictionCNNMulticlass(features_list[0].shape, num_classes=3)
+criterion = nn.CrossEntropyLoss()
+optimizer = optim.Adam(model.parameters(), lr=0.001)
+
+
+load_saved = sys.argv[2] # "last, "best"
+if load_saved != "none":
+    # Load the checkpoint
+    with open(f"model_dumps/{sys.argv[1]}_{load_saved}_model_with_metadata.pkl", "rb") as f:
+        checkpoint = pickle.load(f)
+
+    # Restore model and optimizer state
+    model.load_state_dict(checkpoint["model_state"])
+    optimizer.load_state_dict(checkpoint["optimizer_state"])
+
+    # Restore RNG states
+    torch.set_rng_state(checkpoint["t_rng_st"])
+    np.random.set_state(checkpoint["n_rng_st"])
+    if torch.cuda.is_available() and checkpoint["cuda_rng_st"] is not None:
+        torch.cuda.set_rng_state_all(checkpoint["cuda_rng_st"])
+
+    # Retrieve metadata
+    start_epoch = checkpoint["epoch"] + 1
+    loaded_loss = checkpoint["loss"]
+    loaded_accuracy = checkpoint["accuracy"]
+
+    print(f"Loaded model from epoch {start_epoch} with best validation loss: {loaded_loss:.4f}")
+
+# Initialize lists to track loss
+filename = sys.argv[1] 
+csv_file = f"model_dumps/{filename}.csv"
+
+# Load existing epoch data if the file exists
+if os.path.exists(csv_file):
+    epoch_data = pd.read_csv(csv_file).to_dict(orient="records")
+    print(f"Loaded existing epoch data from {csv_file}.")
+else:
+    epoch_data = []
+    print("No existing epoch data found. Starting fresh.")
+
+
+def update_plot(epoch_data, filename):
+    # Create a DataFrame from epoch data
+    df = pd.DataFrame(epoch_data)
+    df.to_csv(f"model_dumps/{filename}.csv", index=False)  # Save the data to CSV
+
+    # Plot training and validation loss
+    plt.figure(figsize=(10, 8))
+    plt.plot(df['epoch'], df['train_loss'], label="Train Loss", marker="o")
+    plt.plot(df['epoch'], df['val_loss'], label="Validation Loss", marker="o")
+    plt.xlabel("Epoch")
+    plt.ylabel("Loss")
+    plt.title("Loss vs. Epoch")
+    plt.legend()
+    plt.grid(True)
+    plt.savefig(f"model_dumps/{filename}_loss.png")  # Save loss plot
+    plt.close()
+
+    # Plot training and validation accuracy
+    plt.figure(figsize=(10, 8))
+    plt.plot(df['epoch'], df['train_accuracy'], label="Train Accuracy", marker="o", color='green')
+    plt.plot(df['epoch'], df['val_accuracy'], label="Validation Accuracy", marker="o", color='red')
+    plt.xlabel("Epoch")
+    plt.ylabel("Accuracy")
+    plt.title("Accuracy vs. Epoch")
+    plt.legend()
+    plt.grid(True)
+    plt.savefig(f"model_dumps/{filename}_accuracy.png")  # Save accuracy plot
+    plt.close()
+
+
+# Training parameters
+num_epochs = 100
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+model.to(device)
+
+start_epoch = 0
+epochs = 100
+
+best_accuracy = 0.0
+val_accuracy = 0.0
+
+# Train the model
+for epoch in range(start_epoch, epochs):
+    model.train()
+    train_loss = 0.0
+    correct = 0
+    total = 0
+
+    for idx, (features, sex, labels) in enumerate(train_loader):
+        features = features.unsqueeze(1).to(device)  # Add channel dimension
+        sex = sex.to(device)
+        labels = labels.to(device)
+
+        optimizer.zero_grad()
+        outputs = model(features, sex)  # Forward pass
+
+        loss = criterion(outputs, labels)  # CrossEntropyLoss expects raw logits
+        loss.backward()
+        optimizer.step()
+
+        train_loss += loss.item()
+
+        _, predicted = torch.max(outputs, 1)  # Get class predictions
+        total += labels.size(0)
+        correct += (predicted == labels).sum().item()
+
+    train_loss /= len(train_loader)
+    train_accuracy = correct / total
+    print(f"Epoch {epoch+1}/{epochs}, Train Loss: {train_loss:.4f}")
+    print(f"Epoch {epoch+1}/{epochs}, Train Accuracy: {train_accuracy:.4f}")
+
+    # Validation loop
+    model.eval()
+    val_loss = 0.0
+    correct = 0
+    total = 0
+    label_to_class = {0: 'CN', 1: 'AD', 2: 'MCI'}
+    all_preds = []
+    all_labels = []
+
+    with torch.no_grad():
+        for idx, (features, sex, labels) in enumerate(val_loader):
+            features = features.unsqueeze(1).to(device)  # Add channel dimension
+            sex = sex.to(device)
+            labels = labels.to(device)
+
+            outputs = model(features, sex)
+            loss = criterion(outputs, labels)  # CrossEntropyLoss expects raw logits
+            val_loss += loss.item()
+
+            _, predicted = torch.max(outputs, 1)  # Get class predictions
+            total += labels.size(0)
+            correct += (predicted == labels).sum().item()
+
+            all_preds.extend(predicted.cpu().numpy())
+            all_labels.extend(labels.cpu().numpy())
+
+
+    val_loss /= len(val_loader)
+    val_accuracy = correct / total
+
+    print(f"Epoch {epoch+1}/{epochs}, Validation Loss: {val_loss:.4f}")
+    print(f"Epoch {epoch+1}/{epochs}, Validation Accuracy: {val_accuracy:.4f}")
+
+    all_preds_classes = [label_to_class[label] for label in all_preds]
+    all_labels_classes = [label_to_class[label] for label in all_labels]
+    
+    
+
+    checkpoint = {
+        "model_state": model.state_dict(),
+        "optimizer_state": optimizer.state_dict(),
+        "epoch": epoch,
+        "loss": val_loss,
+        "accuracy": val_accuracy,
+        "t_rng_st": torch.get_rng_state(),
+        "n_rng_st": np.random.get_state(),
+        "cuda_rng_st": torch.cuda.get_rng_state_all() if torch.cuda.is_available() else None
+    }
+    with open(f"model_dumps/{sys.argv[1]}_last_model_with_metadata.pkl", "wb") as f:
+        pickle.dump(checkpoint, f)
+    print(f"Last model saved...")
+
+    # Check if validation loss improved
+    if  val_accuracy > best_accuracy:
+        # best_loss = val_loss
+        best_accuracy = val_accuracy
+        print(f"Validation accuracy improved to {val_accuracy:.4f}. Saving the model...")
+        with open(f"model_dumps/{sys.argv[1]}_best_model_with_metadata.pkl", "wb") as f:
+            pickle.dump(checkpoint, f)
+        
+        # Generate confusion matrix
+        cm = confusion_matrix(all_labels_classes, all_preds_classes, labels=['CN', 'AD', 'MCI'])
+        cm_display = sns.heatmap(cm, annot=True, fmt="d", cmap="Blues", xticklabels=['CN', 'AD', 'MCI'], yticklabels=['CN', 'AD', 'MCI'])
+        plt.xlabel("Predicted Class")
+        plt.ylabel("True Class")
+        plt.title(f"Confusion Matrix for Epoch {epoch+1}")
+        plt.savefig(f"model_dumps/{sys.argv[1]}_conf_matrix.png")
+        plt.close()
+        print(f"Confusion matrix for epoch {epoch+1} saved.")
+
+    epoch_data.append({
+        "epoch": epoch + 1,
+        "train_loss": train_loss,
+        "train_accuracy": train_accuracy,
+        "val_loss": val_loss,
+        "val_accuracy": val_accuracy
+    })
+
+    update_plot(epoch_data, sys.argv[1])
+
+
+    
+
+
diff --git a/classify_main_bag.py b/classify_main_bag.py
new file mode 100644
index 0000000..d2156df
--- /dev/null
+++ b/classify_main_bag.py
@@ -0,0 +1,338 @@
+import pandas as pd
+import numpy as np
+import matplotlib.pyplot as plt
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler
+from sklearn.metrics import confusion_matrix
+from sklearn.metrics import accuracy_score
+import torch
+import torch.nn as nn
+import torch.nn.init as init
+import torch.optim as optim
+from tqdm import tqdm
+from torchvision import transforms
+from transformers import ViTFeatureExtractor, ViTModel
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler
+import pickle
+import matplotlib.pyplot as plt
+import sys
+import os
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, Dataset
+import seaborn as sns
+
+def set_random_seed(seed=69420):
+    np.random.seed(seed)
+    torch.manual_seed(seed)
+    torch.cuda.manual_seed(seed)
+    torch.cuda.manual_seed_all(seed)  # For multi-GPU setups
+    torch.backends.cudnn.deterministic = True
+    torch.backends.cudnn.benchmark = False
+
+set_random_seed(69420)
+np.random.seed(69420)
+
+temp_df_1 = pd.read_csv("model_dumps/cnn_mx_lrelu_predicted_ages_train.csv")
+temp_df_2 = pd.read_csv("model_dumps/cnn_mx_lrelu_predicted_ages_val.csv")
+
+# concatenate the two dataframes
+temp_df = pd.concat([temp_df_1, temp_df_2])
+
+# make a column for age gap, absolute difference between predicted and actual age
+
+temp_df["AgeGap"] = abs(temp_df["Predicted_Age"] - temp_df["Age"])
+
+
+# Load the data
+csv_path = "adni_storage/adni_brainrotnet_metadata.csv"
+df = pd.read_csv(csv_path)
+df['filepath'] = df.apply(
+    lambda row: f"adni_storage/ADNI_nii_gz_bias_corrected/I{row['ImageID']}_{row['SubjectID']}.stripped.N4.nii.gz",
+    axis=1
+)
+
+df["AgeGap"] = df["ImageID"].map(temp_df.set_index("ImageID")["AgeGap"])
+
+# convert the Group column to numeric
+df['Group'] = df['Group'].map({'CN': 0, 'AD': 1, 'MCI': 2})
+sex_encoded = df['Sex'].apply(lambda x: 0 if x == 'M' else 1).tolist()
+age_gap_encoded = df['AgeGap'].tolist()
+
+features_list = []
+labels_list = []
+
+for _, row in tqdm(df.iterrows(), total=len(df), desc="Processing images"):
+    filepath = row['filepath']
+    image_title = f"{row['ImageID']}_{row['SubjectID']}"
+
+    # Check if the feature file already exists
+    feature_file_path = f"adni_storage/ADNI_features/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        # scale the features
+        features = StandardScaler().fit_transform(features)
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Group'])  # Add the corresponding age label
+
+# Convert the lists to numpy arrays
+features = np.array(features_list)
+labels = np.array(labels_list)
+
+#apply standard scaling to the features and age gap_encoded, sex_encoded
+scaler = StandardScaler()
+sex_encoded = scaler.fit_transform(np.array(sex_encoded).reshape(-1, 1)).flatten()
+age_gap_encoded = scaler.fit_transform(np.array(age_gap_encoded).reshape(-1, 1)).flatten()
+
+
+
+# Custom Dataset
+class ADNIDataset(Dataset):
+    def __init__(self, features_list, sex_list, age_gap_list, labels_list):
+        self.features = features_list
+        self.sex = sex_list
+        self.labels = labels_list
+        self.age_gap = age_gap_list
+   
+
+    def __len__(self):
+        return len(self.features)
+
+    def __getitem__(self, idx):
+        return (
+            torch.tensor(self.features[idx], dtype=torch.float32),
+            torch.tensor(self.sex[idx], dtype=torch.float32),
+            torch.tensor(self.age_gap[idx], dtype=torch.float32),
+            torch.tensor(self.labels[idx], dtype=torch.long),
+        )
+    
+# Create Dataset and DataLoader
+dataset = ADNIDataset(features_list, sex_encoded, age_gap_encoded, labels)
+train_size = int(0.8 * len(dataset))
+val_size = len(dataset) - train_size
+train_dataset, val_dataset = torch.utils.data.random_split(dataset, [train_size, val_size])
+
+train_loader = DataLoader(train_dataset, batch_size=1, shuffle=False)
+val_loader = DataLoader(val_dataset, batch_size=1, shuffle=False)
+
+
+import importlib
+
+module_name = sys.argv[1]  # Example: "my_model"
+class_name = "AgePredictionCNNMulticlass"  # The class you want to import
+
+try:
+    # Dynamically import the module
+    module = importlib.import_module(module_name)
+    
+    # Dynamically get the class
+    AgePredictionCNNMulticlass = getattr(module, class_name)
+    
+    print(f"Successfully imported {class_name} from {module_name}.")
+
+except ImportError:
+    print(f"Module {module_name} could not be imported.")
+except AttributeError:
+    print(f"{class_name} does not exist in {module_name}.")
+
+model = AgePredictionCNNMulticlass(features_list[0].shape, num_classes=3)
+criterion = nn.CrossEntropyLoss()
+optimizer = optim.Adam(model.parameters(), lr=0.001)
+
+
+load_saved = sys.argv[2] # "last, "best"
+if load_saved != "none":
+    # Load the checkpoint
+    with open(f"model_dumps/{sys.argv[1]}_{load_saved}_model_with_metadata.pkl", "rb") as f:
+        checkpoint = pickle.load(f)
+
+    # Restore model and optimizer state
+    model.load_state_dict(checkpoint["model_state"])
+    optimizer.load_state_dict(checkpoint["optimizer_state"])
+
+    # Restore RNG states
+    torch.set_rng_state(checkpoint["t_rng_st"])
+    np.random.set_state(checkpoint["n_rng_st"])
+    if torch.cuda.is_available() and checkpoint["cuda_rng_st"] is not None:
+        torch.cuda.set_rng_state_all(checkpoint["cuda_rng_st"])
+
+    # Retrieve metadata
+    start_epoch = checkpoint["epoch"] + 1
+    loaded_loss = checkpoint["loss"]
+    loaded_accuracy = checkpoint["accuracy"]
+
+    print(f"Loaded model from epoch {start_epoch} with best validation loss: {loaded_loss:.4f}")
+
+# Initialize lists to track loss
+filename = sys.argv[1] 
+csv_file = f"model_dumps/{filename}.csv"
+
+# Load existing epoch data if the file exists
+if os.path.exists(csv_file):
+    epoch_data = pd.read_csv(csv_file).to_dict(orient="records")
+    print(f"Loaded existing epoch data from {csv_file}.")
+else:
+    epoch_data = []
+    print("No existing epoch data found. Starting fresh.")
+
+
+def update_plot(epoch_data, filename):
+    # Create a DataFrame from epoch data
+    df = pd.DataFrame(epoch_data)
+    df.to_csv(f"model_dumps/{filename}.csv", index=False)  # Save the data to CSV
+
+    # Plot training and validation loss
+    plt.figure(figsize=(10, 8))
+    plt.plot(df['epoch'], df['train_loss'], label="Train Loss", marker="o")
+    plt.plot(df['epoch'], df['val_loss'], label="Validation Loss", marker="o")
+    plt.xlabel("Epoch")
+    plt.ylabel("Loss")
+    plt.title("Loss vs. Epoch")
+    plt.legend()
+    plt.grid(True)
+    plt.savefig(f"model_dumps/{filename}_loss.png")  # Save loss plot
+    plt.close()
+
+    # Plot training and validation accuracy
+    plt.figure(figsize=(10, 8))
+    plt.plot(df['epoch'], df['train_accuracy'], label="Train Accuracy", marker="o", color='green')
+    plt.plot(df['epoch'], df['val_accuracy'], label="Validation Accuracy", marker="o", color='red')
+    plt.xlabel("Epoch")
+    plt.ylabel("Accuracy")
+    plt.title("Accuracy vs. Epoch")
+    plt.legend()
+    plt.grid(True)
+    plt.savefig(f"model_dumps/{filename}_accuracy.png")  # Save accuracy plot
+    plt.close()
+
+
+# Training parameters
+num_epochs = 100
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+model.to(device)
+
+start_epoch = 0
+epochs = 100
+
+best_accuracy = 0.0
+val_accuracy = 0.0
+
+# Train the model
+for epoch in range(start_epoch, epochs):
+    model.train()
+    train_loss = 0.0
+    correct = 0
+    total = 0
+
+    for idx, (features, sex, age_gap, labels) in enumerate(train_loader):
+        features = features.unsqueeze(1).to(device)  # Add channel dimension
+        sex = sex.to(device)
+        labels = labels.to(device)
+        age_gap = age_gap.to(device)
+
+        optimizer.zero_grad()
+        outputs = model(features, sex, age_gap)
+
+        loss = criterion(outputs, labels)  # CrossEntropyLoss expects raw logits
+        loss.backward()
+        optimizer.step()
+
+        train_loss += loss.item()
+
+        _, predicted = torch.max(outputs, 1)  # Get class predictions
+        total += labels.size(0)
+        correct += (predicted == labels).sum().item()
+
+    train_loss /= len(train_loader)
+    train_accuracy = correct / total
+    print(f"Epoch {epoch+1}/{epochs}, Train Loss: {train_loss:.4f}")
+    print(f"Epoch {epoch+1}/{epochs}, Train Accuracy: {train_accuracy:.4f}")
+
+    # Validation loop
+    model.eval()
+    val_loss = 0.0
+    correct = 0
+    total = 0
+    label_to_class = {0: 'CN', 1: 'AD', 2: 'MCI'}
+    all_preds = []
+    all_labels = []
+
+    with torch.no_grad():
+        for idx, (features, sex, age_gap, labels) in enumerate(val_loader):
+            features = features.unsqueeze(1).to(device)  # Add channel dimension
+            sex = sex.to(device)
+            labels = labels.to(device)
+            age_gap = age_gap.to(device)
+
+            outputs = model(features, sex, age_gap)
+            loss = criterion(outputs, labels)  # CrossEntropyLoss expects raw logits
+            val_loss += loss.item()
+
+            _, predicted = torch.max(outputs, 1)  # Get class predictions
+            total += labels.size(0)
+            correct += (predicted == labels).sum().item()
+
+            all_preds.extend(predicted.cpu().numpy())
+            all_labels.extend(labels.cpu().numpy())
+
+
+    val_loss /= len(val_loader)
+    val_accuracy = correct / total
+
+    print(f"Epoch {epoch+1}/{epochs}, Validation Loss: {val_loss:.4f}")
+    print(f"Epoch {epoch+1}/{epochs}, Validation Accuracy: {val_accuracy:.4f}")
+
+    all_preds_classes = [label_to_class[label] for label in all_preds]
+    all_labels_classes = [label_to_class[label] for label in all_labels]
+    
+    
+
+    checkpoint = {
+        "model_state": model.state_dict(),
+        "optimizer_state": optimizer.state_dict(),
+        "epoch": epoch,
+        "loss": val_loss,
+        "accuracy": val_accuracy,
+        "t_rng_st": torch.get_rng_state(),
+        "n_rng_st": np.random.get_state(),
+        "cuda_rng_st": torch.cuda.get_rng_state_all() if torch.cuda.is_available() else None
+    }
+    with open(f"model_dumps/{sys.argv[1]}_last_model_with_metadata.pkl", "wb") as f:
+        pickle.dump(checkpoint, f)
+    print(f"Last model saved...")
+
+    # Check if validation loss improved
+    if  val_accuracy > best_accuracy:
+        # best_loss = val_loss
+        best_accuracy = val_accuracy
+        print(f"Validation accuracy improved to {val_accuracy:.4f}. Saving the model...")
+        with open(f"model_dumps/{sys.argv[1]}_best_model_with_metadata.pkl", "wb") as f:
+            pickle.dump(checkpoint, f)
+        
+        # Generate confusion matrix
+        cm = confusion_matrix(all_labels_classes, all_preds_classes, labels=['CN', 'AD', 'MCI'])
+        cm_display = sns.heatmap(cm, annot=True, fmt="d", cmap="Blues", xticklabels=['CN', 'AD', 'MCI'], yticklabels=['CN', 'AD', 'MCI'])
+        plt.xlabel("Predicted Class")
+        plt.ylabel("True Class")
+        plt.title(f"Confusion Matrix for Epoch {epoch+1}")
+        plt.savefig(f"model_dumps/{sys.argv[1]}_conf_matrix.png")
+        plt.close()
+        print(f"Confusion matrix for epoch {epoch+1} saved.")
+
+    epoch_data.append({
+        "epoch": epoch + 1,
+        "train_loss": train_loss,
+        "train_accuracy": train_accuracy,
+        "val_loss": val_loss,
+        "val_accuracy": val_accuracy
+    })
+
+    update_plot(epoch_data, sys.argv[1])
+
+
+    
+
+
diff --git a/cnn_gn.py b/cnn_gn.py
new file mode 100644
index 0000000..5cc6b74
--- /dev/null
+++ b/cnn_gn.py
@@ -0,0 +1,246 @@
+# import torch
+# import torch.nn as nn
+# import torch.nn.functional as F
+
+# def conv_block(in_ch, out_ch, kernel_size, padding, groups):
+#     return nn.Sequential(
+#         nn.Conv2d(in_ch, out_ch, kernel_size=kernel_size, padding=padding),
+#         nn.GroupNorm(num_groups=groups, num_channels=out_ch),
+#         nn.SiLU(),
+#         nn.MaxPool2d(kernel_size=2, stride=1),
+#         nn.Dropout2d(p=0.1),
+#     )
+
+# class AgePredictionCNN(nn.Module):
+#     def __init__(self,input_shape=None):
+#         super(AgePredictionCNN, self).__init__()
+
+#         # Convolutional blocks with GroupNorm
+#         self.block1 = conv_block(1, 16, kernel_size=(10, 60), padding=(5, 30), groups=4)
+#         self.block2 = conv_block(16, 32, kernel_size=(5, 15), padding=(2, 7), groups=8)
+#         self.block3 = conv_block(32, 64, kernel_size=(2, 6), padding=(1, 3), groups=16)
+
+#         # Adaptive pooling to get fixed-size output regardless of input dims
+#         self.global_pool = nn.AdaptiveAvgPool2d((1, 1))
+
+#         # Fully connected layers
+#         self.fc1 = nn.Linear(64, 128)
+#         self.fc1_gn = nn.GroupNorm(num_groups=8, num_channels=128)
+#         self.dropout = nn.Dropout(p=0.2)
+
+#         # Final layer: +1 for sex input
+#         self.fc2 = nn.Linear(128 + 1, 1)
+
+#         self._init_weights()
+
+#     def _init_weights(self):
+#         for m in self.modules():
+#             if isinstance(m, (nn.Conv2d, nn.Linear)):
+#                 nn.init.kaiming_normal_(m.weight, nonlinearity='relu')
+#                 if m.bias is not None:
+#                     nn.init.constant_(m.bias, 0)
+
+#     def forward(self, x, sex):
+#         # x: [B, 1, H, W], sex: [B]
+#         x = self.block1(x)
+#         x = self.block2(x)
+#         x = self.block3(x)
+
+#         x = self.global_pool(x)      # Shape: [B, 64, 1, 1]
+#         x = torch.flatten(x, 1)      # Shape: [B, 64]
+
+#         x = self.fc1(x)
+#         x = self.fc1_gn(x)
+#         x = F.silu(x)
+#         x = self.dropout(x)
+
+#         x = torch.cat([x, sex.unsqueeze(1)], dim=1)  # Concatenate sex
+#         out = self.fc2(x)
+
+#         return out
+
+# import torch
+# import torch.nn as nn
+# import torch.optim as optim
+
+# class AgePredictionCNN(nn.Module):
+#     def __init__(self, input_shape):
+#         super(AgePredictionCNN, self).__init__()
+#         # Define convolutional and pooling layers with typical kernel sizes and GroupNorm
+#         self.conv1 = nn.Conv2d(1, 16, kernel_size=5, stride=1, padding=1)
+#         self.gn1 = nn.GroupNorm(num_groups=4, num_channels=16)
+#         self.pool1 = nn.MaxPool2d(kernel_size=2, stride=2)
+
+#         self.conv2 = nn.Conv2d(16, 32, kernel_size=5, stride=1, padding=1)
+#         self.gn2 = nn.GroupNorm(num_groups=4, num_channels=32)
+#         self.pool2 = nn.MaxPool2d(kernel_size=2, stride=2)
+
+#         self.conv3 = nn.Conv2d(32, 64, kernel_size=5, stride=1, padding=1)
+#         self.gn3 = nn.GroupNorm(num_groups=8, num_channels=64)
+#         self.pool3 = nn.MaxPool2d(kernel_size=2, stride=2)
+#         # Adaptive pooling to get fixed-size output regardless of input dims
+#         self.flatten = nn.Flatten()
+
+#         # Fully connected layers
+#         self.fc1 = None  # Placeholder to be initialized dynamically
+#         self.fc1_ln = None  # LayerNorm after fc1
+#         self.fc2 = nn.Linear(512, 128)
+#         self.fc2_ln = nn.LayerNorm(128)
+#         self.dropout = nn.Dropout(p=0.4)
+#         self.fc3 = nn.Linear(129, 1)  # Adding 1 for the `Sex` input
+
+#         self.swish = nn.SiLU()
+
+#         self.initialize_fc1(input_shape)
+#         self.initialize_weights()
+
+#     def initialize_fc1(self, input_shape):
+#         # Create a sample input to pass through the convolutional layers
+#         sample_input = torch.zeros(1, *input_shape)
+#         x = self.conv1(sample_input)
+#         x = self.swish(x)
+#         x = self.pool1(x)
+#         x = self.conv2(x)
+#         x = self.swish(x)
+#         x = self.pool2(x)
+#         x = self.conv3(x)
+#         x = self.swish(x)
+#         x = self.pool3(x)
+#         flattened_size = x.numel()
+#         self.fc1 = nn.Linear(flattened_size, 512)
+#         self.fc1_ln = nn.LayerNorm(512)
+
+#     def initialize_weights(self):
+#         # Initialize weights for all layers
+#         for m in self.modules():
+#             if isinstance(m, nn.Conv2d):
+#                 nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
+#                 if m.bias is not None:
+#                     nn.init.constant_(m.bias, 0)
+#             elif isinstance(m, (nn.GroupNorm, nn.LayerNorm)):
+#                 nn.init.constant_(m.weight, 1)
+#                 nn.init.constant_(m.bias, 0)
+#             elif isinstance(m, nn.Linear):
+#                 nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
+#                 nn.init.constant_(m.bias, 0)
+
+#     def forward(self, x, sex):
+#         x = self.conv1(x)
+#         x = self.gn1(x)
+#         x = self.swish(x)
+#         x = self.pool1(x)
+
+#         x = self.conv2(x)
+#         x = self.gn2(x)
+#         x = self.swish(x)
+#         x = self.pool2(x)
+
+#         x = self.conv3(x)
+#         x = self.gn3(x)
+#         x = self.swish(x)
+#         x = self.pool3(x)
+
+#         x = self.flatten(x)
+
+#         if self.fc1 is None:
+#             raise ValueError("fc1 layer has not been initialized. Call `initialize_fc1` with the input shape.")
+
+#         x = self.fc1(x)
+#         x = self.fc1_ln(x)
+#         x = self.swish(x)
+#         x = self.dropout(x)
+
+#         x = self.fc2(x)
+#         x = self.fc2_ln(x)
+#         x = self.swish(x)
+#         x = self.dropout(x)
+
+#         x = torch.cat((x, sex.unsqueeze(1)), dim=1)
+#         x = self.fc3(x)
+
+#         return x
+
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+class AgePredictionCNN(nn.Module):
+    def __init__(self, input_shape):
+        super(AgePredictionCNN, self).__init__()
+
+        self.conv1 = nn.Conv2d(1, 16, kernel_size=7, stride=1, padding=1)
+        self.gn1 = nn.GroupNorm(num_groups=4, num_channels=16)
+        self.pool1 = nn.MaxPool2d(kernel_size=2, stride=2)
+        self.drop2d1 = nn.Dropout2d(p=0.3)
+
+        self.conv2 = nn.Conv2d(16, 32, kernel_size=7, stride=1, padding=1)
+        self.gn2 = nn.GroupNorm(num_groups=4, num_channels=32)
+        self.pool2 = nn.MaxPool2d(kernel_size=2, stride=2)
+        self.drop2d2 = nn.Dropout2d(p=0.3)
+
+        self.conv3 = nn.Conv2d(32, 64, kernel_size=7, stride=1, padding=1)
+        self.gn3 = nn.GroupNorm(num_groups=8, num_channels=64)
+        self.pool3 = nn.MaxPool2d(kernel_size=2, stride=2)
+        self.drop2d3 = nn.Dropout2d(p=0.3)
+
+        self.flatten = nn.Flatten()
+
+        # Fully connected layers
+        self.fc1 = None
+        self.fc1_ln = None
+        self.fc2 = nn.Linear(256, 128)
+        self.fc2_ln = nn.LayerNorm(128)
+        self.dropout = nn.Dropout(p=0.5)
+        self.fc3 = nn.Linear(129, 1)  # 128 + 1 (sex)
+
+        self.swish = nn.SiLU()
+
+        self.initialize_fc1(input_shape)
+        self.initialize_weights()
+
+    def initialize_fc1(self, input_shape):
+        with torch.no_grad():
+            sample_input = torch.zeros(1, *input_shape)
+            x = self.forward_conv(sample_input)
+            flattened_size = x.numel()
+        self.fc1 = nn.Linear(flattened_size, 256)
+        self.fc1_ln = nn.LayerNorm(256)
+
+    def initialize_weights(self):
+        for m in self.modules():
+            if isinstance(m, nn.Conv2d):
+                nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
+                if m.bias is not None:
+                    nn.init.constant_(m.bias, 0)
+            elif isinstance(m, (nn.GroupNorm, nn.LayerNorm)):
+                nn.init.constant_(m.weight, 1)
+                nn.init.constant_(m.bias, 0)
+            elif isinstance(m, nn.Linear):
+                nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
+                nn.init.constant_(m.bias, 0)
+
+    def forward_conv(self, x):
+        x = self.pool1(self.swish(self.gn1(self.conv1(x))))
+        x = self.drop2d1(x)
+
+        x = self.pool2(self.swish(self.gn2(self.conv2(x))))
+        x = self.drop2d2(x)
+
+        x = self.pool3(self.swish(self.gn3(self.conv3(x))))
+        x = self.drop2d3(x)
+
+        return self.flatten(x)
+
+    def forward(self, x, sex):
+        x = self.forward_conv(x)
+
+        x = self.swish(self.fc1_ln(self.fc1(x)))
+        x = self.dropout(x)
+
+        x = self.swish(self.fc2_ln(self.fc2(x)))
+        x = self.dropout(x)
+
+        x = torch.cat((x, sex.unsqueeze(1)), dim=1)
+        x = self.fc3(x)
+
+        return x
diff --git a/cnn_mx.py b/cnn_mx.py
index 3f9b421..fc57de4 100644
--- a/cnn_mx.py
+++ b/cnn_mx.py
@@ -25,7 +25,7 @@ def __init__(self, input_shape):
         self.fc1_bn = None  # Placeholder for batch normalization after fc1
         self.fc2 = nn.Linear(512, 128)
         self.fc2_bn = nn.LayerNorm(128)
-        self.dropout = nn.Dropout(p=0.1)  # Dropout with 10% probability
+        self.dropout = nn.Dropout(p=0.3)  # Dropout with 10% probability
         self.fc3 = nn.Linear(129, 1)  # Adding 1 for the `Sex` input
 
         self.relu = nn.ReLU()
@@ -78,3 +78,91 @@ def forward(self, x, sex):
 
         return x
 
+# import torch
+# import torch.nn as nn
+# import torch.optim as optim
+# from torch.utils.data import DataLoader, Dataset
+
+# class AgePredictionCNN(nn.Module):
+#     def __init__(self, input_shape):
+#         super(AgePredictionCNN, self).__init__()
+
+#         # Define convolutional and pooling layers
+#         self.conv1 = nn.Conv2d(1, 1, kernel_size=(10, 60), stride=1)
+#         self.pool1 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+#         self.conv2 = nn.Conv2d(1, 1, kernel_size=(5, 15), stride=1)
+#         self.pool2 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+#         self.conv3 = nn.Conv2d(1, 1, kernel_size=(2, 6), stride=1)
+#         self.pool3 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+#         self.flatten = nn.Flatten()
+
+#         # Fully connected layers (fc1 dimensions are calculated dynamically)
+#         self.fc1 = None  # Placeholder to be initialized dynamically
+#         self.fc1_bn = None  # Placeholder for batch normalization after fc1
+#         self.fc2 = nn.Linear(512, 128)
+#         self.fc2_bn = nn.LayerNorm(128)
+#         self.dropout = nn.Dropout(p=0.3)  # Dropout with 10% probability
+#         self.fc3 = nn.Linear(129, 1)  # Adding 1 for the `Sex` input
+
+#         self.relu = nn.ReLU()
+#         self.initialize_fc1(input_shape)
+#         self.init_weights()  # Initialize weights
+
+#     def initialize_fc1(self, input_shape):
+#         # Create a sample input to pass through the convolutional layers
+#         sample_input = torch.zeros(1, *input_shape)
+#         x = self.conv1(sample_input)
+#         x = self.relu(x)
+#         x = self.pool1(x)  # Apply pooling
+#         x = self.conv2(x)
+#         x = self.relu(x)
+#         x = self.pool2(x)  # Apply pooling
+#         x = self.conv3(x)
+#         x = self.relu(x)
+#         x = self.pool3(x)  # Apply pooling
+#         flattened_size = x.numel()  # Total number of elements after flattening
+#         self.fc1 = nn.Linear(flattened_size, 512)
+#         self.fc1_bn = nn.LayerNorm(512)  # Initialize batch normalization for fc1
+
+#     def init_weights(self):
+#         # Apply xavier initialization to all the layers
+#         for layer in self.modules():
+#             if isinstance(layer, nn.Conv2d) or isinstance(layer, nn.Linear):
+#                 nn.init.xavier_uniform_(layer.weight)
+#                 if layer.bias is not None:
+#                     layer.bias.data.fill_(0.01)
+                
+
+#     def forward(self, x, sex):
+#         x = self.conv1(x)
+#         x = self.relu(x)
+#         x = self.pool1(x)  # Apply pooling
+#         x = self.conv2(x)
+#         x = self.relu(x)
+#         x = self.pool2(x)  # Apply pooling
+#         x = self.conv3(x)
+#         x = self.relu(x)
+#         x = self.pool3(x)  # Apply pooling
+#         x = self.flatten(x)
+
+#         if self.fc1 is None:
+#             raise ValueError("fc1 layer has not been initialized. Call `initialize_fc1` with the input shape.")
+
+#         x = self.fc1(x)
+#         x = self.fc1_bn(x)  # Apply batch normalization
+#         x = self.relu(x)
+#         x = self.dropout(x)  # Apply dropout
+
+#         x = self.fc2(x)
+#         x = self.fc2_bn(x)  # Apply batch normalization
+#         x = self.relu(x)
+#         x = self.dropout(x)  # Apply dropout
+
+#         # Concatenate `Sex` input
+#         x = torch.cat((x, sex.unsqueeze(1)), dim=1)
+#         x = self.fc3(x)
+
+#         return x
diff --git a/cnn_mx_att.py b/cnn_mx_att.py
new file mode 100644
index 0000000..7d72a7e
--- /dev/null
+++ b/cnn_mx_att.py
@@ -0,0 +1,95 @@
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, Dataset
+
+
+class AgePredictionCNN(nn.Module):
+    def __init__(self, input_shape):
+        super(AgePredictionCNN, self).__init__()
+
+        # Define convolutional and pooling layers
+        self.conv1 = nn.Conv2d(1, 1, kernel_size=(10, 60), stride=1)
+        self.pool1 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+        self.conv2 = nn.Conv2d(1, 1, kernel_size=(5, 15), stride=1)
+        self.pool2 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+        self.conv3 = nn.Conv2d(1, 1, kernel_size=(2, 6), stride=1)
+        self.pool3 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+        self.flatten = nn.Flatten()
+
+        # Fully connected layers (fc1 dimensions are calculated dynamically)
+        self.fc1 = None  # Placeholder to be initialized dynamically
+        self.fc1_bn = None  # Placeholder for batch normalization after fc1
+        self.self_attention_512 = nn.MultiheadAttention(embed_dim=512, num_heads=8)
+        
+        self.fc2 = nn.Linear(512, 128)
+        self.fc2_bn = nn.LayerNorm(128)
+        self.self_attention_128 = nn.MultiheadAttention(embed_dim=128, num_heads=4)
+
+        self.dropout = nn.Dropout(p=0.1)  # Dropout with 10% probability
+        self.fc3 = nn.Linear(129, 1)  # Adding 1 for the `Sex` input
+
+        self.relu = nn.ReLU()
+        self.initialize_fc1(input_shape)
+
+    def initialize_fc1(self, input_shape):
+        # Create a sample input to pass through the convolutional layers
+        sample_input = torch.zeros(1, *input_shape)
+        x = self.conv1(sample_input)
+        x = self.relu(x)
+        x = self.pool1(x)  # Apply pooling
+        x = self.conv2(x)
+        x = self.relu(x)
+        x = self.pool2(x)  # Apply pooling
+        x = self.conv3(x)
+        x = self.relu(x)
+        x = self.pool3(x)  # Apply pooling
+        flattened_size = x.numel()  # Total number of elements after flattening
+        self.fc1 = nn.Linear(flattened_size, 512)
+        self.fc1_bn = nn.LayerNorm(512)  # Initialize batch normalization for fc1
+
+    def forward(self, x, sex):
+        x = self.conv1(x)
+        x = self.relu(x)
+        x = self.pool1(x)  # Apply pooling
+        x = self.conv2(x)
+        x = self.relu(x)
+        x = self.pool2(x)  # Apply pooling
+        x = self.conv3(x)
+        x = self.relu(x)
+        x = self.pool3(x)  # Apply pooling
+        x = self.flatten(x)
+
+        if self.fc1 is None:
+            raise ValueError("fc1 layer has not been initialized. Call `initialize_fc1` with the input shape.")
+
+        x = self.fc1(x)
+        x = self.fc1_bn(x)  # Apply batch normalization
+        x = self.relu(x)
+
+        # # Self-attention for 512-dimensional features
+        # x = x.unsqueeze(0)  # Add sequence dimension for attention
+        # x, _ = self.self_attention_512(x, x, x)
+        # x = x.squeeze(0)  # Remove sequence dimension
+
+        x = self.dropout(x)  # Apply dropout
+
+        x = self.fc2(x)
+        x = self.fc2_bn(x)  # Apply batch normalization
+        x = self.relu(x)
+
+        # # Self-attention for 128-dimensional features
+        # x = x.unsqueeze(0)  # Add sequence dimension for attention
+        # x, _ = self.self_attention_128(x, x, x)
+        # x = x.squeeze(0)  # Remove sequence dimension
+
+        x = self.dropout(x)  # Apply dropout
+
+        # Concatenate `Sex` input
+        x = torch.cat((x, sex.unsqueeze(1)), dim=1)
+        x = self.fc3(x)
+
+        return x
diff --git a/cnn_mx_att_bl.py b/cnn_mx_att_bl.py
new file mode 100644
index 0000000..a7b60b2
--- /dev/null
+++ b/cnn_mx_att_bl.py
@@ -0,0 +1,106 @@
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+class AttentionBlock(nn.Module):
+    def __init__(self, channels, reduction=16):
+        super(AttentionBlock, self).__init__()
+        self.global_avg_pool = nn.AdaptiveAvgPool2d(1)  # Global Average Pooling
+        self.fc1 = nn.Linear(channels, channels // reduction, bias=False)
+        self.relu = nn.ReLU()
+        self.fc2 = nn.Linear(channels // reduction, channels, bias=False)
+        self.sigmoid = nn.Sigmoid()
+
+    def forward(self, x):
+        b, c, _, _ = x.size()  # Batch size, Channels, Height, Width
+        y = self.global_avg_pool(x).view(b, c)  # Squeeze spatial dimensions
+        y = self.fc1(y)
+        y = self.relu(y)
+        y = self.fc2(y)
+        y = self.sigmoid(y).view(b, c, 1, 1)  # Reshape to match input dimensions
+        return x * y  # Scale input by attention weights
+
+
+class AgePredictionCNN(nn.Module):
+    def __init__(self, input_shape):
+        super(AgePredictionCNN, self).__init__()
+
+        # Define convolutional layers with 1 channel
+        self.conv1 = nn.Conv2d(1, 1, kernel_size=(10, 60), stride=1, padding=1)
+        self.pool1 = nn.MaxPool2d(kernel_size=2, stride=2)
+
+        self.conv2 = nn.Conv2d(1, 1, kernel_size=(5, 15), stride=1, padding=1)
+        self.pool2 = nn.MaxPool2d(kernel_size=2, stride=2)
+
+        self.conv3 = nn.Conv2d(1, 1, kernel_size=(2, 6), stride=1, padding=1)
+        self.pool3 = nn.MaxPool2d(kernel_size=2, stride=2)
+
+        # Attention blocks
+        self.attention1 = AttentionBlock(channels=1)
+        self.attention2 = AttentionBlock(channels=1)
+        self.attention3 = AttentionBlock(channels=1)
+
+        self.flatten = nn.Flatten()
+
+        # Fully connected layers
+        self.fc1 = None  # Placeholder to initialize dynamically
+        self.fc1_bn = None  # Placeholder for batch normalization after fc1
+        self.fc2 = nn.Linear(512, 128)
+        self.fc2_bn = nn.LayerNorm(128)
+        self.dropout = nn.Dropout(p=0.1)  # Dropout with 10% probability
+        self.fc3 = nn.Linear(129, 1)  # Adding 1 for the `Sex` input
+
+        self.relu = nn.ReLU()
+        self.initialize_fc1(input_shape)
+
+    def initialize_fc1(self, input_shape):
+        sample_input = torch.zeros(1, *input_shape)
+        x = self.conv1(sample_input)
+        x = self.relu(x)
+        x = self.pool1(x)
+        x = self.conv2(x)
+        x = self.relu(x)
+        x = self.pool2(x)
+        x = self.conv3(x)
+        x = self.relu(x)
+        x = self.pool3(x)
+        flattened_size = x.numel()
+        self.fc1 = nn.Linear(flattened_size, 512)
+        self.fc1_bn = nn.LayerNorm(512)
+
+    def forward(self, x, sex):
+        # First convolution block with attention
+        x = self.conv1(x)
+        x = self.attention1(x)  # Apply attention
+        x = self.relu(x)
+        x = self.pool1(x)
+
+        # Second convolution block with attention
+        x = self.conv2(x)
+        x = self.attention2(x)  # Apply attention
+        x = self.relu(x)
+        x = self.pool2(x)
+
+        # Third convolution block with attention
+        x = self.conv3(x)
+        x = self.attention3(x)  # Apply attention
+        x = self.relu(x)
+        x = self.pool3(x)
+
+        # Flatten and pass through fully connected layers
+        x = self.flatten(x)
+        x = self.fc1(x)
+        x = self.fc1_bn(x)
+        x = self.relu(x)
+        x = self.dropout(x)
+
+        x = self.fc2(x)
+        x = self.fc2_bn(x)
+        x = self.relu(x)
+        x = self.dropout(x)
+
+        # Concatenate `Sex` input
+        x = torch.cat((x, sex.unsqueeze(1)), dim=1)
+        x = self.fc3(x)
+
+        return x
diff --git a/cnn_mx_att_gate.py b/cnn_mx_att_gate.py
new file mode 100644
index 0000000..07833f5
--- /dev/null
+++ b/cnn_mx_att_gate.py
@@ -0,0 +1,110 @@
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, Dataset
+
+class SpatialAttention(nn.Module):
+    def __init__(self, kernel_size=7):
+        super(SpatialAttention, self).__init__()
+        # Ensure kernel size is odd for consistent padding
+        assert kernel_size % 2 == 1, "Kernel size must be odd"
+        self.conv = nn.Conv2d(1, 1, kernel_size=kernel_size, padding=kernel_size // 2, bias=False)
+        self.sigmoid = nn.Sigmoid()
+
+    def forward(self, x):
+        # Average and Max pooling across the channel dimension
+        avg_pool = torch.mean(x, dim=1, keepdim=True)
+        max_pool, _ = torch.max(x, dim=1, keepdim=True)
+        # Concatenate along channel dimension
+        pooled = avg_pool + max_pool
+        # Apply convolution and sigmoid activation
+        attention_map = self.sigmoid(self.conv(pooled))
+        # Multiply input with attention map
+        return x * attention_map
+
+
+class AgePredictionCNN(nn.Module):
+    def __init__(self, input_shape):
+        super(AgePredictionCNN, self).__init__()
+
+        # Define convolutional and pooling layers
+        self.conv1 = nn.Conv2d(1, 1, kernel_size=(10, 60), stride=1)
+        self.pool1 = nn.MaxPool2d(kernel_size=2, stride=1)
+        self.sa1 = SpatialAttention(kernel_size=7)  # Add spatial attention after first layer
+
+        self.conv2 = nn.Conv2d(1, 1, kernel_size=(5, 15), stride=1)
+        self.pool2 = nn.MaxPool2d(kernel_size=2, stride=1)
+        self.sa2 = SpatialAttention(kernel_size=5)  # Add spatial attention after second layer
+
+        self.conv3 = nn.Conv2d(1, 1, kernel_size=(2, 6), stride=1)
+        self.pool3 = nn.MaxPool2d(kernel_size=2, stride=1)
+        self.sa3 = SpatialAttention(kernel_size=3)  # Add spatial attention after third layer
+
+        self.flatten = nn.Flatten()
+
+        # Fully connected layers
+        self.fc1 = None  # Placeholder to be initialized dynamically
+        self.fc1_bn = None  # Placeholder for batch normalization after fc1
+        self.fc2 = nn.Linear(512, 128)
+        self.fc2_bn = nn.LayerNorm(128)
+        self.dropout = nn.Dropout(p=0.3)  # Dropout with 30% probability
+        self.fc3 = nn.Linear(129, 1)  # Adding 1 for the `Sex` input
+
+        self.relu = nn.ReLU()
+        self.initialize_fc1(input_shape)
+
+    def initialize_fc1(self, input_shape):
+        # Create a sample input to pass through the convolutional layers
+        sample_input = torch.zeros(1, *input_shape)
+        x = self.conv1(sample_input)
+        x = self.relu(x)
+        x = self.pool1(x)
+        x = self.sa1(x)  # Apply spatial attention
+        x = self.conv2(x)
+        x = self.relu(x)
+        x = self.pool2(x)
+        x = self.sa2(x)  # Apply spatial attention
+        x = self.conv3(x)
+        x = self.relu(x)
+        x = self.pool3(x)
+        x = self.sa3(x)  # Apply spatial attention
+        flattened_size = x.numel()  # Total number of elements after flattening
+        self.fc1 = nn.Linear(flattened_size, 512)
+        self.fc1_bn = nn.LayerNorm(512)  # Initialize batch normalization for fc1
+
+    def forward(self, x, sex):
+        x = self.conv1(x)
+        x = self.relu(x)
+        x = self.pool1(x)
+        x = self.sa1(x)  # Apply spatial attention
+
+        x = self.conv2(x)
+        x = self.relu(x)
+        x = self.pool2(x)
+        x = self.sa2(x)  # Apply spatial attention
+
+        x = self.conv3(x)
+        x = self.relu(x)
+        x = self.pool3(x)
+        x = self.sa3(x)  # Apply spatial attention
+
+        x = self.flatten(x)
+
+        if self.fc1 is None:
+            raise ValueError("fc1 layer has not been initialized. Call `initialize_fc1` with the input shape.")
+
+        x = self.fc1(x)
+        x = self.fc1_bn(x)  # Apply batch normalization
+        x = self.relu(x)
+        x = self.dropout(x)  # Apply dropout
+
+        x = self.fc2(x)
+        x = self.fc2_bn(x)  # Apply batch normalization
+        x = self.relu(x)
+        x = self.dropout(x)  # Apply dropout
+
+        # Concatenate `Sex` input
+        x = torch.cat((x, sex.unsqueeze(1)), dim=1)
+        x = self.fc3(x)
+
+        return x
diff --git a/cnn_mx_att_gate2.py b/cnn_mx_att_gate2.py
new file mode 100644
index 0000000..237f1b7
--- /dev/null
+++ b/cnn_mx_att_gate2.py
@@ -0,0 +1,119 @@
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, Dataset
+
+
+class AttentionGate(nn.Module):
+    def __init__(self, F_g, F_l, F_int):
+        super(AttentionGate, self).__init__()
+        self.W_g = nn.Conv2d(F_g, F_int, kernel_size=1, stride=1, padding=0, bias=True)
+        self.W_x = nn.Conv2d(F_l, F_int, kernel_size=1, stride=1, padding=0, bias=True)
+        self.psi = nn.Conv2d(F_int, 1, kernel_size=1, stride=1, padding=0, bias=True)
+        self.relu = nn.ReLU(inplace=True)
+        self.sigmoid = nn.Sigmoid()
+
+    def forward(self, g, x):
+        # Resize g to match x's spatial dimensions
+        g = nn.functional.interpolate(g, size=x.shape[2:], mode='bilinear', align_corners=False)
+
+        g1 = self.W_g(g)
+        x1 = self.W_x(x)
+        psi = self.relu(g1 + x1)
+        psi = self.sigmoid(self.psi(psi))
+        return x * psi
+
+
+class AgePredictionCNN(nn.Module):
+    def __init__(self, input_shape):
+        super(AgePredictionCNN, self).__init__()
+
+        # Define convolutional and pooling layers
+        self.conv1 = nn.Conv2d(1, 1, kernel_size=(10, 60), stride=1)
+        self.pool1 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+        self.conv2 = nn.Conv2d(1, 1, kernel_size=(5, 15), stride=1)
+        self.pool2 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+        self.conv3 = nn.Conv2d(1, 1, kernel_size=(2, 6), stride=1)
+        self.pool3 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+        # Attention Gates for spatial refinement
+        self.ag1 = AttentionGate(F_g=1, F_l=1, F_int=1)
+        self.ag2 = AttentionGate(F_g=1, F_l=1, F_int=1)
+
+        self.flatten = nn.Flatten()
+
+        # Fully connected layers
+        self.fc1 = None  # Placeholder to be initialized dynamically
+        self.fc1_bn = None  # Placeholder for batch normalization after fc1
+        self.fc2 = nn.Linear(512, 128)
+        self.fc2_bn = nn.LayerNorm(128)
+        self.dropout = nn.Dropout(p=0.3)  # Dropout with 30% probability
+        self.fc3 = nn.Linear(129, 1)  # Adding 1 for the `Sex` input
+
+        self.relu = nn.ReLU()
+        self.initialize_fc1(input_shape)
+
+    def initialize_fc1(self, input_shape):
+        # Create a sample input to pass through the convolutional layers
+        sample_input = torch.zeros(1, *input_shape)
+        x1 = self.conv1(sample_input)
+        x1 = self.relu(x1)
+        x1 = self.pool1(x1)
+
+        x2 = self.conv2(x1)
+        x2 = self.relu(x2)
+        x2 = self.pool2(x2)
+
+        x3 = self.conv3(x2)
+        x3 = self.relu(x3)
+        x3 = self.pool3(x3)
+
+        # Attention Gates
+        x2 = self.ag1(g=x3, x=x2)  # First attention gate
+        x1 = self.ag2(g=x2, x=x1)  # Second attention gate
+
+        flattened_size = x1.numel()  # Total number of elements after flattening
+        self.fc1 = nn.Linear(flattened_size, 512)
+        self.fc1_bn = nn.LayerNorm(512)  # Initialize batch normalization for fc1
+
+    def forward(self, x, sex):
+        # Initial convolution layers
+        x1 = self.conv1(x)
+        x1 = self.relu(x1)
+        x1 = self.pool1(x1)
+
+        x2 = self.conv2(x1)
+        x2 = self.relu(x2)
+        x2 = self.pool2(x2)
+
+        x3 = self.conv3(x2)
+        x3 = self.relu(x3)
+        x3 = self.pool3(x3)
+
+        # Attention Gates
+        x2 = self.ag1(g=x3, x=x2)  # First attention gate
+        x1 = self.ag2(g=x2, x=x1)  # Second attention gate
+
+        # Flatten and fully connected layers
+        x = self.flatten(x1)
+
+        if self.fc1 is None:
+            raise ValueError("fc1 layer has not been initialized. Call `initialize_fc1` with the input shape.")
+
+        x = self.fc1(x)
+        x = self.fc1_bn(x)  # Apply batch normalization
+        x = self.relu(x)
+        x = self.dropout(x)  # Apply dropout
+
+        x = self.fc2(x)
+        x = self.fc2_bn(x)  # Apply batch normalization
+        x = self.relu(x)
+        x = self.dropout(x)  # Apply dropout
+
+        # Concatenate `Sex` input
+        x = torch.cat((x, sex.unsqueeze(1)), dim=1)
+        x = self.fc3(x)
+
+        return x
diff --git a/cnn_mx_att_res.py b/cnn_mx_att_res.py
new file mode 100644
index 0000000..0744931
--- /dev/null
+++ b/cnn_mx_att_res.py
@@ -0,0 +1,117 @@
+# import torch
+# import torch.nn as nn
+# import torch.optim as optim
+# from torch.utils.data import DataLoader, Dataset
+
+# class ResidualAttentionGate(nn.Module):
+#     def __init__(self, F_g, F_l, F_int):
+#         super(ResidualAttentionGate, self).__init__()
+#         self.W_g = nn.Conv2d(F_g, F_int, kernel_size=1, stride=1, padding=0, bias=True)
+#         self.W_x = nn.Conv2d(F_l, F_int, kernel_size=1, stride=1, padding=0, bias=True)
+#         self.psi = nn.Conv2d(F_int, 1, kernel_size=1, stride=1, padding=0, bias=True)
+#         self.relu = nn.ReLU(inplace=True)
+#         self.sigmoid = nn.Sigmoid()
+
+#     def forward(self, g, x):
+#         # Process inputs through attention
+#         g1 = self.W_g(g)
+#         x1 = self.W_x(x)
+#         psi = self.relu(g1 + x1)
+#         psi = self.sigmoid(self.psi(psi))
+        
+#         # Apply attention to the input feature map `x`
+#         attention = x * psi
+
+#         # Add residual connection (original input `x` + attention-weighted input)
+#         output = x + attention
+#         return output
+
+
+# class AgePredictionCNN(nn.Module):
+#     def __init__(self, input_shape):
+#         super(AgePredictionCNN, self).__init__()
+
+#         # Define convolutional and pooling layers
+#         self.conv1 = nn.Conv2d(1, 1, kernel_size=(10, 60), stride=1)
+#         self.pool1 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+#         self.conv2 = nn.Conv2d(1, 1, kernel_size=(5, 15), stride=1)
+#         self.pool2 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+#         self.conv3 = nn.Conv2d(1, 1, kernel_size=(2, 6), stride=1)
+#         self.pool3 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+#         # Residual Attention Gates
+#         self.att_gate1 = ResidualAttentionGate(F_g=1, F_l=1, F_int=1)
+#         self.att_gate2 = ResidualAttentionGate(F_g=1, F_l=1, F_int=1)
+
+#         self.flatten = nn.Flatten()
+
+#         # Fully connected layers (fc1 dimensions are calculated dynamically)
+#         self.fc1 = None  # Placeholder to be initialized dynamically
+#         self.fc1_bn = None  # Placeholder for batch normalization after fc1
+#         self.fc2 = nn.Linear(512, 128)
+#         self.fc2_bn = nn.LayerNorm(128)
+#         self.dropout = nn.Dropout(p=0.3)  # Dropout with 10% probability
+#         self.fc3 = nn.Linear(129, 1)  # Adding 1 for the `Sex` input
+
+#         self.relu = nn.ReLU()
+#         self.initialize_fc1(input_shape)
+
+#     def initialize_fc1(self, input_shape):
+#         # Create a sample input to pass through the convolutional layers
+#         sample_input = torch.zeros(1, *input_shape)
+#         x = self.conv1(sample_input)
+#         x = self.relu(x)
+#         x = self.pool1(x)
+#         x = self.conv2(x)
+#         x = self.relu(x)
+#         x = self.pool2(x)
+#         x = self.conv3(x)
+#         x = self.relu(x)
+#         x = self.pool3(x)
+        
+#         # Pass through attention gates
+#         x = self.att_gate1(g=x, x=x)
+#         x = self.att_gate2(g=x, x=x)
+
+#         flattened_size = x.numel()
+#         self.fc1 = nn.Linear(flattened_size, 512)
+#         self.fc1_bn = nn.LayerNorm(512)
+
+#     def forward(self, x, sex):
+#         x = self.conv1(x)
+#         x = self.relu(x)
+#         x = self.pool1(x)
+#         x = self.conv2(x)
+#         x = self.relu(x)
+#         x = self.pool2(x)
+#         x = self.conv3(x)
+#         x = self.relu(x)
+#         x = self.pool3(x)
+
+#         # Pass through residual attention gates
+#         x = self.att_gate1(g=x, x=x)
+#         x = self.att_gate2(g=x, x=x)
+
+#         x = self.flatten(x)
+
+#         if self.fc1 is None:
+#             raise ValueError("fc1 layer has not been initialized. Call `initialize_fc1` with the input shape.")
+
+#         x = self.fc1(x)
+#         x = self.fc1_bn(x)
+#         x = self.relu(x)
+#         x = self.dropout(x)
+
+#         x = self.fc2(x)
+#         x = self.fc2_bn(x)
+#         x = self.relu(x)
+#         x = self.dropout(x)
+
+#         # Concatenate `Sex` input
+#         x = torch.cat((x, sex.unsqueeze(1)), dim=1)
+#         x = self.fc3(x)
+
+#         return x
+
diff --git a/cnn_mx_lrelu.py b/cnn_mx_lrelu.py
new file mode 100644
index 0000000..6887451
--- /dev/null
+++ b/cnn_mx_lrelu.py
@@ -0,0 +1,80 @@
+
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, Dataset
+
+class AgePredictionCNN(nn.Module):
+    def __init__(self, input_shape):
+        super(AgePredictionCNN, self).__init__()
+
+        # Define convolutional and pooling layers
+        self.conv1 = nn.Conv2d(1, 1, kernel_size=(10, 60), stride=1)
+        self.pool1 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+        self.conv2 = nn.Conv2d(1, 1, kernel_size=(5, 15), stride=1)
+        self.pool2 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+        self.conv3 = nn.Conv2d(1, 1, kernel_size=(2, 6), stride=1)
+        self.pool3 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+        self.flatten = nn.Flatten()
+
+        # Fully connected layers (fc1 dimensions are calculated dynamically)
+        self.fc1 = None  # Placeholder to be initialized dynamically
+        self.fc1_bn = None  # Placeholder for batch normalization after fc1
+        self.fc2 = nn.Linear(512, 128)
+        self.fc2_bn = nn.LayerNorm(128)
+        self.dropout = nn.Dropout(p=0.1)  # Dropout with 10% probability
+        self.fc3 = nn.Linear(129, 1)  # Adding 1 for the `Sex` input
+        
+        self.lrelu = nn.LeakyReLU()
+        self.initialize_fc1(input_shape)
+
+    def initialize_fc1(self, input_shape):
+        # Create a sample input to pass through the convolutional layers
+        sample_input = torch.zeros(1, *input_shape)
+        x = self.conv1(sample_input)
+        x = self.lrelu(x)
+        x = self.pool1(x)  # Apply pooling
+        x = self.conv2(x)
+        x = self.lrelu(x)
+        x = self.pool2(x)  # Apply pooling
+        x = self.conv3(x)
+        x = self.lrelu(x)
+        x = self.pool3(x)  # Apply pooling
+        flattened_size = x.numel()  # Total number of elements after flattening
+        self.fc1 = nn.Linear(flattened_size, 512)
+        self.fc1_bn = nn.LayerNorm(512)  # Initialize batch normalization for fc1
+
+    def forward(self, x, sex):
+        x = self.conv1(x)
+        x = self.lrelu(x)
+        x = self.pool1(x)  # Apply pooling
+        x = self.conv2(x)
+        x = self.lrelu(x)
+        x = self.pool2(x)  # Apply pooling
+        x = self.conv3(x)
+        x = self.lrelu(x)
+        x = self.pool3(x)  # Apply pooling
+        x = self.flatten(x)
+
+        if self.fc1 is None:
+            raise ValueError("fc1 layer has not been initialized. Call `initialize_fc1` with the input shape.")
+
+        x = self.fc1(x)
+        x = self.fc1_bn(x)  # Apply batch normalization
+        x = self.lrelu(x)
+        x = self.dropout(x)  # Apply dropout
+
+        x = self.fc2(x)
+        x = self.fc2_bn(x)  # Apply batch normalization
+        x = self.lrelu(x)
+        x = self.dropout(x)  # Apply dropout
+
+        # Concatenate `Sex` input
+        x = torch.cat((x, sex.unsqueeze(1)), dim=1)
+        x = self.fc3(x)
+
+        return x
+
diff --git a/cnn_mx_multi_att.py b/cnn_mx_multi_att.py
new file mode 100644
index 0000000..2310f68
--- /dev/null
+++ b/cnn_mx_multi_att.py
@@ -0,0 +1,110 @@
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, Dataset
+
+class MultiScaleAttention(nn.Module):
+    def __init__(self, in_channels):
+        super(MultiScaleAttention, self).__init__()
+        # Convolutions with different kernel sizes for multi-scale feature extraction
+        self.conv1 = nn.Conv2d(in_channels, 1, kernel_size=3, padding=1)
+        self.conv2 = nn.Conv2d(in_channels, 1, kernel_size=5, padding=2)
+        self.conv3 = nn.Conv2d(in_channels, 1, kernel_size=7, padding=3)
+        # Combine multi-scale features
+        self.sigmoid = nn.Sigmoid()
+
+    def forward(self, x):
+        # Extract multi-scale features
+        feat1 = self.conv1(x)
+        feat2 = self.conv2(x)
+        feat3 = self.conv3(x)
+        # Aggregate features and apply sigmoid for attention
+        attention = self.sigmoid(feat1 + feat2 + feat3)
+        # Refine the input features with the attention map
+        return x * attention
+
+class AgePredictionCNN(nn.Module):
+    def __init__(self, input_shape):
+        super(AgePredictionCNN, self).__init__()
+
+        # Define convolutional and pooling layers
+        self.conv1 = nn.Conv2d(1, 1, kernel_size=(10, 60), stride=1)
+        self.pool1 = nn.MaxPool2d(kernel_size=2, stride=1)
+        self.att1 = MultiScaleAttention(1)  # Multi-scale attention after conv1
+
+        self.conv2 = nn.Conv2d(1, 1, kernel_size=(5, 15), stride=1)
+        self.pool2 = nn.MaxPool2d(kernel_size=2, stride=1)
+        self.att2 = MultiScaleAttention(1)  # Multi-scale attention after conv2
+
+        self.conv3 = nn.Conv2d(1, 1, kernel_size=(2, 6), stride=1)
+        self.pool3 = nn.MaxPool2d(kernel_size=2, stride=1)
+        self.att3 = MultiScaleAttention(1)  # Multi-scale attention after conv3
+
+        self.flatten = nn.Flatten()
+
+        # Fully connected layers
+        self.fc1 = None  # Placeholder to be initialized dynamically
+        self.fc1_bn = None  # Placeholder for batch normalization after fc1
+        self.fc2 = nn.Linear(512, 128)
+        self.fc2_bn = nn.LayerNorm(128)
+        self.dropout = nn.Dropout(p=0.3)  # Dropout
+        self.fc3 = nn.Linear(129, 1)  # Adding 1 for the `Sex` input
+
+        self.relu = nn.ReLU()
+        self.initialize_fc1(input_shape)
+
+    def initialize_fc1(self, input_shape):
+        # Create a sample input to pass through the convolutional layers
+        sample_input = torch.zeros(1, *input_shape)
+        x = self.conv1(sample_input)
+        x = self.relu(x)
+        x = self.pool1(x)
+        x = self.att1(x)  # Apply attention after first pooling
+        x = self.conv2(x)
+        x = self.relu(x)
+        x = self.pool2(x)
+        x = self.att2(x)  # Apply attention after second pooling
+        x = self.conv3(x)
+        x = self.relu(x)
+        x = self.pool3(x)
+        x = self.att3(x)  # Apply attention after third pooling
+        flattened_size = x.numel()
+        self.fc1 = nn.Linear(flattened_size, 512)
+        self.fc1_bn = nn.LayerNorm(512)
+
+    def forward(self, x, sex):
+        x = self.conv1(x)
+        x = self.relu(x)
+        x = self.pool1(x)
+        x = self.att1(x)  # Apply attention
+
+        x = self.conv2(x)
+        x = self.relu(x)
+        x = self.pool2(x)
+        x = self.att2(x)  # Apply attention
+
+        x = self.conv3(x)
+        x = self.relu(x)
+        x = self.pool3(x)
+        x = self.att3(x)  # Apply attention
+
+        x = self.flatten(x)
+
+        if self.fc1 is None:
+            raise ValueError("fc1 layer has not been initialized. Call `initialize_fc1` with the input shape.")
+
+        x = self.fc1(x)
+        x = self.fc1_bn(x)
+        x = self.relu(x)
+        x = self.dropout(x)
+
+        x = self.fc2(x)
+        x = self.fc2_bn(x)
+        x = self.relu(x)
+        x = self.dropout(x)
+
+        # Concatenate `Sex` input
+        x = torch.cat((x, sex.unsqueeze(1)), dim=1)
+        x = self.fc3(x)
+
+        return x
diff --git a/cnn_mx_spd.py b/cnn_mx_spd.py
new file mode 100644
index 0000000..fc6e737
--- /dev/null
+++ b/cnn_mx_spd.py
@@ -0,0 +1,97 @@
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, Dataset
+
+
+class SpatialDropout2d(nn.Dropout2d):
+    def forward(self, x):
+        if not self.training or self.p == 0:
+            return x
+        mask = torch.ones((x.size(0), x.size(1), 1, 1), device=x.device, dtype=x.dtype)
+        mask = nn.functional.dropout(mask, self.p, self.training, inplace=True)
+        return x * mask
+
+
+class AgePredictionCNN(nn.Module):
+    def __init__(self, input_shape):
+        super(AgePredictionCNN, self).__init__()
+
+        # Define convolutional and pooling layers
+        self.conv1 = nn.Conv2d(1, 1, kernel_size=(10, 60), stride=1)
+        self.pool1 = nn.MaxPool2d(kernel_size=2, stride=1)
+        self.spatial_dropout1 = SpatialDropout2d(p=0.2)  # Added SpatialDropout2d after conv1
+
+        self.conv2 = nn.Conv2d(1, 1, kernel_size=(5, 15), stride=1)
+        self.pool2 = nn.MaxPool2d(kernel_size=2, stride=1)
+        self.spatial_dropout2 = SpatialDropout2d(p=0.2)  # Added SpatialDropout2d after conv2
+
+        self.conv3 = nn.Conv2d(1, 1, kernel_size=(2, 6), stride=1)
+        self.pool3 = nn.MaxPool2d(kernel_size=2, stride=1)
+        self.spatial_dropout3 = SpatialDropout2d(p=0.2)  # Added SpatialDropout2d after conv3
+
+        self.flatten = nn.Flatten()
+
+        # Fully connected layers (fc1 dimensions are calculated dynamically)
+        self.fc1 = None  # Placeholder to be initialized dynamically
+        self.fc1_bn = None  # Placeholder for batch normalization after fc1
+        self.fc2 = nn.Linear(512, 128)
+        self.fc2_bn = nn.LayerNorm(128)
+        self.dropout = nn.Dropout(p=0.3)  # Dropout with 10% probability
+        self.fc3 = nn.Linear(129, 1)  # Adding 1 for the `Sex` input
+
+        self.swish = nn.SiLU()  # Swish activation function
+        self.initialize_fc1(input_shape)
+
+    def initialize_fc1(self, input_shape):
+        # Create a sample input to pass through the convolutional layers
+        sample_input = torch.zeros(1, *input_shape)
+        x = self.conv1(sample_input)
+        x = self.swish(x)
+        x = self.pool1(x)  # Apply pooling
+        x = self.conv2(x)
+        x = self.swish(x)
+        x = self.pool2(x)  # Apply pooling
+        x = self.conv3(x)
+        x = self.swish(x)
+        x = self.pool3(x)  # Apply pooling
+        flattened_size = x.numel()  # Total number of elements after flattening
+        self.fc1 = nn.Linear(flattened_size, 512)
+        self.fc1_bn = nn.LayerNorm(512)  # Initialize batch normalization for fc1
+
+    def forward(self, x, sex):
+        x = self.conv1(x)
+        x = self.swish(x)
+        x = self.pool1(x)  # Apply pooling
+        x = self.spatial_dropout1(x)  # Apply SpatialDropout2d
+
+        x = self.conv2(x)
+        x = self.swish(x)
+        x = self.pool2(x)  # Apply pooling
+        x = self.spatial_dropout2(x)  # Apply SpatialDropout2d
+
+        x = self.conv3(x)
+        x = self.swish(x)
+        x = self.pool3(x)  # Apply pooling
+        x = self.spatial_dropout3(x)  # Apply SpatialDropout2d
+
+        x = self.flatten(x)
+
+        if self.fc1 is None:
+            raise ValueError("fc1 layer has not been initialized. Call `initialize_fc1` with the input shape.")
+
+        x = self.fc1(x)
+        x = self.fc1_bn(x)  # Apply batch normalization
+        x = self.swish(x)
+        x = self.dropout(x)  # Apply dropout
+
+        x = self.fc2(x)
+        x = self.fc2_bn(x)  # Apply batch normalization
+        x = self.swish(x)
+        x = self.dropout(x)  # Apply dropout
+
+        # Concatenate `Sex` input
+        x = torch.cat((x, sex.unsqueeze(1)), dim=1)
+        x = self.fc3(x)
+
+        return x
diff --git a/cnn_mx_sw.py b/cnn_mx_sw.py
new file mode 100644
index 0000000..00d3196
--- /dev/null
+++ b/cnn_mx_sw.py
@@ -0,0 +1,114 @@
+
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+import torch.optim as optim
+from torch.utils.data import DataLoader, Dataset
+
+
+# best one so far
+
+import torch
+import torch.nn as nn
+import torch.optim as optim
+
+class AgePredictionCNN(nn.Module):
+    def __init__(self, input_shape):
+        super(AgePredictionCNN, self).__init__()
+
+        # Define convolutional and pooling layers
+        self.conv1 = nn.Conv2d(1, 1, kernel_size=(10, 60), stride=1)
+        self.pool1 = nn.MaxPool2d(kernel_size=2, stride=1)
+        self.bn1 = nn.BatchNorm2d(1)
+
+        self.conv2 = nn.Conv2d(1, 1, kernel_size=(5, 15), stride=1)
+        self.pool2 = nn.MaxPool2d(kernel_size=2, stride=1)
+        self.bn2 = nn.BatchNorm2d(1)
+
+        self.conv3 = nn.Conv2d(1, 1, kernel_size=(2, 6), stride=1)
+        self.pool3 = nn.MaxPool2d(kernel_size=2, stride=1)
+        self.bn3 = nn.BatchNorm2d(1)
+
+        self.flatten = nn.Flatten()
+
+        # Fully connected layers
+        self.fc1 = None  # Placeholder to be initialized dynamically
+        self.fc1_bn = None  # Placeholder for batch normalization after fc1
+        self.fc2 = nn.Linear(512, 128)
+        self.fc2_bn = nn.LayerNorm(128)
+        self.dropout = nn.Dropout(p=0.1)
+        self.fc3 = nn.Linear(129, 1)  # Adding 1 for the `Sex` input
+
+        self.swish = nn.SiLU()
+
+        self.initialize_fc1(input_shape)
+        self.initialize_weights()  # Call the custom weight initialization
+
+    def initialize_fc1(self, input_shape):
+        # Create a sample input to pass through the convolutional layers
+        sample_input = torch.zeros(1, *input_shape)
+        x = self.conv1(sample_input)
+        x = self.swish(x)
+        x = self.pool1(x)
+        x = self.conv2(x)
+        x = self.swish(x)
+        x = self.pool2(x)
+        x = self.conv3(x)
+        x = self.swish(x)
+        x = self.pool3(x)
+        flattened_size = x.numel()
+        self.fc1 = nn.Linear(flattened_size, 512)
+        self.fc1_bn = nn.LayerNorm(512)
+
+    def initialize_weights(self):
+        # Initialize weights for all layers
+        for m in self.modules():
+            if isinstance(m, nn.Conv2d):
+                nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
+                #nn.init.xavier_normal_(m.weight)
+                if m.bias is not None:
+                    nn.init.constant_(m.bias, 0)
+            elif isinstance(m, nn.BatchNorm2d) or isinstance(m, nn.LayerNorm):
+                nn.init.constant_(m.weight, 1)
+                nn.init.constant_(m.bias, 0)
+            elif isinstance(m, nn.Linear):
+                #nn.init.xavier_normal_(m.weight)
+                nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
+                nn.init.constant_(m.bias, 0)
+
+    def forward(self, x, sex):
+        x = self.conv1(x)
+        x = self.bn1(x)
+        x = self.swish(x)
+        x = self.pool1(x)
+
+        x = self.conv2(x)
+        x = self.bn2(x)
+        x = self.swish(x)
+        x = self.pool2(x)
+
+        x = self.conv3(x)
+        x = self.bn3(x)
+        x = self.swish(x)
+        x = self.pool3(x)
+
+        x = self.flatten(x)
+
+        if self.fc1 is None:
+            raise ValueError("fc1 layer has not been initialized. Call `initialize_fc1` with the input shape.")
+
+        x = self.fc1(x)
+        x = self.fc1_bn(x)
+        x = self.swish(x)
+        x = self.dropout(x)
+
+        x = self.fc2(x)
+        x = self.fc2_bn(x)
+        x = self.swish(x)
+        x = self.dropout(x)
+
+        # Concatenate `Sex` input
+        x = torch.cat((x, sex.unsqueeze(1)), dim=1)
+        x = self.fc3(x)
+
+        return x
diff --git a/cnn_mx_vl.py b/cnn_mx_vl.py
new file mode 100644
index 0000000..5e81c2c
--- /dev/null
+++ b/cnn_mx_vl.py
@@ -0,0 +1,87 @@
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, Dataset
+
+class AgePredictionCNN(nn.Module):
+    def __init__(self, input_shape):
+        super(AgePredictionCNN, self).__init__()
+
+        # Define convolutional and pooling layers
+        self.conv1 = nn.Conv2d(1, 1, kernel_size=(10, 85), stride=1)
+        self.pool1 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+        self.conv2 = nn.Conv2d(1, 1, kernel_size=(5, 20), stride=1)
+        self.pool2 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+        self.conv3 = nn.Conv2d(1, 1, kernel_size=(2, 10), stride=1)
+        self.pool3 = nn.MaxPool2d(kernel_size=2, stride=1)
+
+        self.flatten = nn.Flatten()
+
+        # Fully connected layers (fc1 dimensions are calculated dynamically)
+        self.fc1 = None  # Placeholder to be initialized dynamically
+        self.fc1_bn = None  # Placeholder for batch normalization after fc1
+        self.fc2 = nn.Linear(1024, 512)
+        self.fc2_bn = nn.LayerNorm(512)
+        self.fc3 = nn.Linear(512, 128)
+        self.fc3_bn = nn.LayerNorm(128)
+        self.dropout = nn.Dropout(p=0.1)  # Dropout with 10% probability
+        self.fc4 = nn.Linear(129, 1)  # Adding 1 for the `Sex` input
+
+        self.relu = nn.ReLU()
+        self.initialize_fc1(input_shape)
+
+    def initialize_fc1(self, input_shape):
+        # Create a sample input to pass through the convolutional layers
+        sample_input = torch.zeros(1, *input_shape)
+        x = self.conv1(sample_input)
+        x = self.relu(x)
+        x = self.pool1(x)  # Apply pooling
+        x = self.conv2(x)
+        x = self.relu(x)
+        x = self.pool2(x)  # Apply pooling
+        x = self.conv3(x)
+        x = self.relu(x)
+        x = self.pool3(x)  # Apply pooling
+     
+        flattened_size = x.numel()  # Total number of elements after flattening
+        self.fc1 = nn.Linear(flattened_size, 1024)
+        self.fc1_bn = nn.LayerNorm(1024)  # Initialize batch normalization for fc1
+
+    def forward(self, x, sex):
+        x = self.conv1(x)
+        x = self.relu(x)
+        x = self.pool1(x)  # Apply pooling
+        x = self.conv2(x)
+        x = self.relu(x)
+        x = self.pool2(x)  # Apply pooling
+        x = self.conv3(x)
+        x = self.relu(x)
+        x = self.pool3(x)  # Apply pooling
+    
+        x = self.flatten(x)
+
+        if self.fc1 is None:
+            raise ValueError("fc1 layer has not been initialized. Call `initialize_fc1` with the input shape.")
+
+        x = self.fc1(x)
+        x = self.fc1_bn(x)  # Apply batch normalization
+        x = self.relu(x)
+        x = self.dropout(x)  # Apply dropout
+
+        x = self.fc2(x)
+        x = self.fc2_bn(x)  # Apply batch normalization
+        x = self.relu(x)
+        x = self.dropout(x)  # Apply dropout
+
+        x = self.fc3(x)
+        x = self.fc3_bn(x)  # Apply batch normalization
+        x = self.relu(x)
+        x = self.dropout(x)  # Apply dropout
+
+        # Concatenate `Sex` input
+        x = torch.cat((x, sex.unsqueeze(1)), dim=1)
+        x = self.fc4(x)
+
+        return x
diff --git a/combined_dataset.csv b/combined_dataset.csv
new file mode 100644
index 0000000..75c718e
--- /dev/null
+++ b/combined_dataset.csv
@@ -0,0 +1,2232 @@
+ImageID,Sex,Age,filepath
+ADNI1283849,F,62.4,adni_storage/ADNI_nii_gz_bias_corrected/I1283849_305_S_6845.stripped.N4.nii.gz
+ADNI308453,M,61.3,adni_storage/ADNI_nii_gz_bias_corrected/I308453_073_S_4795.stripped.N4.nii.gz
+fcon1000Milwaukee_sub61779,F,56.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub61779.stripped.N4.nii.gz
+fcon1000Beijing_sub67844,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub67844.stripped.N4.nii.gz
+ADNI997517,F,57.9,adni_storage/ADNI_nii_gz_bias_corrected/I997517_131_S_6170.stripped.N4.nii.gz
+DLBS0028440_session_1,F,34.25,dlbs_storage/DLBS_bias_corrected/0028440_session_1.stripped.N4.nii.gz
+sub-CC520279,F,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520279.stripped.N4.nii.gz
+sub-CC610653,M,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610653.stripped.N4.nii.gz
+fcon1000AnnArbor_sub18698,M,29.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub18698.stripped.N4.nii.gz
+DLBS0028358_session_1,M,66.63,dlbs_storage/DLBS_bias_corrected/0028358_session_1.stripped.N4.nii.gz
+sub-CC420587,F,49.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420587.stripped.N4.nii.gz
+sub-CC520147,F,61.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520147.stripped.N4.nii.gz
+fcon1000Cambridge_sub13216,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub13216.stripped.N4.nii.gz
+sub-CC410091,F,57.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410091.stripped.N4.nii.gz
+ADNI341152,M,60.6,adni_storage/ADNI_nii_gz_bias_corrected/I341152_128_S_4599.stripped.N4.nii.gz
+fcon1000Newark_sub41006,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub41006.stripped.N4.nii.gz
+DLBS0028376_session_1,F,70.6,dlbs_storage/DLBS_bias_corrected/0028376_session_1.stripped.N4.nii.gz
+ADNI363016,F,61.4,adni_storage/ADNI_nii_gz_bias_corrected/I363016_036_S_4894.stripped.N4.nii.gz
+sub-CC510076,M,62.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510076.stripped.N4.nii.gz
+fcon1000Cambridge_sub04270,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub04270.stripped.N4.nii.gz
+ADNI1469766,F,58.6,adni_storage/ADNI_nii_gz_bias_corrected/I1469766_073_S_6670.stripped.N4.nii.gz
+fcon1000Berlin_sub91966,F,27.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub91966.stripped.N4.nii.gz
+fcon1000Cambridge_sub15432,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub15432.stripped.N4.nii.gz
+fcon1000AnnArbor_sub96621,M,15.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub96621.stripped.N4.nii.gz
+fcon1000Beijing_sub58614,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub58614.stripped.N4.nii.gz
+sub-CC420149,M,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420149.stripped.N4.nii.gz
+fcon1000AnnArbor_sub90127,M,77.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub90127.stripped.N4.nii.gz
+DLBS0028633_session_1,M,46.44,dlbs_storage/DLBS_bias_corrected/0028633_session_1.stripped.N4.nii.gz
+DLBS0028565_session_1,M,80.84,dlbs_storage/DLBS_bias_corrected/0028565_session_1.stripped.N4.nii.gz
+sub-CC320202,M,45.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320202.stripped.N4.nii.gz
+ADNI999074,M,62.0,adni_storage/ADNI_nii_gz_bias_corrected/I999074_130_S_6329.stripped.N4.nii.gz
+ADNI1490106,M,57.8,adni_storage/ADNI_nii_gz_bias_corrected/I1490106_037_S_6993.stripped.N4.nii.gz
+ADNI10453857,F,58.6,adni_storage/ADNI_nii_gz_bias_corrected/I10453857_037_S_10086.stripped.N4.nii.gz
+ADNI313319,F,56.5,adni_storage/ADNI_nii_gz_bias_corrected/I313319_128_S_4607.stripped.N4.nii.gz
+fcon1000Milwaukee_sub76378,F,52.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub76378.stripped.N4.nii.gz
+fcon1000AnnArbor_sub16960,M,13.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub16960.stripped.N4.nii.gz
+sub-CC520078,M,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520078.stripped.N4.nii.gz
+fcon1000Cleveland_sub80263,F,44.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub80263.stripped.N4.nii.gz
+DLBS0028497_session_1,F,49.13,dlbs_storage/DLBS_bias_corrected/0028497_session_1.stripped.N4.nii.gz
+sub-CC620793,M,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620793.stripped.N4.nii.gz
+fcon1000Cleveland_sub12330,F,54.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub12330.stripped.N4.nii.gz
+ADNI341366,M,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I341366_116_S_4625.stripped.N4.nii.gz
+ADNI90756,F,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I90756_051_S_1072.stripped.N4.nii.gz
+ADNI308459,M,61.3,adni_storage/ADNI_nii_gz_bias_corrected/I308459_073_S_4795.stripped.N4.nii.gz
+sub-CC520477,M,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520477.stripped.N4.nii.gz
+DLBS0028620_session_1,F,70.66,dlbs_storage/DLBS_bias_corrected/0028620_session_1.stripped.N4.nii.gz
+ADNI98971,M,63.9,adni_storage/ADNI_nii_gz_bias_corrected/I98971_016_S_0769.stripped.N4.nii.gz
+ADNI1639219,M,57.4,adni_storage/ADNI_nii_gz_bias_corrected/I1639219_011_S_7112.stripped.N4.nii.gz
+sub-CC110069,F,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110069.stripped.N4.nii.gz
+fcon1000Beijing_sub39725,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub39725.stripped.N4.nii.gz
+DLBS0028585_session_1,F,31.4,dlbs_storage/DLBS_bias_corrected/0028585_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub87745,M,14.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub87745.stripped.N4.nii.gz
+fcon1000Newark_sub86204,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub86204.stripped.N4.nii.gz
+sub-CC510323,F,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510323.stripped.N4.nii.gz
+fcon1000Cambridge_sub62383,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub62383.stripped.N4.nii.gz
+sub-CC120234,F,24.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120234.stripped.N4.nii.gz
+ADNI1495816,M,59.3,adni_storage/ADNI_nii_gz_bias_corrected/I1495816_003_S_7010.stripped.N4.nii.gz
+sub-CC310008,M,45.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310008.stripped.N4.nii.gz
+sub-CC320553,F,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320553.stripped.N4.nii.gz
+DLBS0028450_session_1,F,44.99,dlbs_storage/DLBS_bias_corrected/0028450_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub18758,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub18758.stripped.N4.nii.gz
+ADNI984775,F,59.1,adni_storage/ADNI_nii_gz_bias_corrected/I984775_305_S_6313.stripped.N4.nii.gz
+sub-CC721377,M,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721377.stripped.N4.nii.gz
+sub-CC510050,F,59.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510050.stripped.N4.nii.gz
+ADNI1568731,F,56.3,adni_storage/ADNI_nii_gz_bias_corrected/I1568731_067_S_7063.stripped.N4.nii.gz
+fcon1000Cambridge_sub54846,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub54846.stripped.N4.nii.gz
+sub-CC223286,M,36.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC223286.stripped.N4.nii.gz
+sub-CC121106,F,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121106.stripped.N4.nii.gz
+ADNI10888787,F,59.2,adni_storage/ADNI_nii_gz_bias_corrected/I10888787_005_S_10101.stripped.N4.nii.gz
+ADNI1315300,F,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I1315300_305_S_6313.stripped.N4.nii.gz
+fcon1000Beijing_sub20765,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub20765.stripped.N4.nii.gz
+fcon1000Beijing_sub73245,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub73245.stripped.N4.nii.gz
+fcon1000Beijing_sub80927,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub80927.stripped.N4.nii.gz
+sub-CC220372,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220372.stripped.N4.nii.gz
+fcon1000Cleveland_sub99664,F,57.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub99664.stripped.N4.nii.gz
+fcon1000Milwaukee_sub91468,M,44.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub91468.stripped.N4.nii.gz
+fcon1000Cambridge_sub83683,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub83683.stripped.N4.nii.gz
+sub-CC520254,F,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520254.stripped.N4.nii.gz
+DLBS0028422_session_1,F,32.91,dlbs_storage/DLBS_bias_corrected/0028422_session_1.stripped.N4.nii.gz
+sub-CC320160,F,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320160.stripped.N4.nii.gz
+DLBS0028612_session_1,F,67.61,dlbs_storage/DLBS_bias_corrected/0028612_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub74386,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub74386.stripped.N4.nii.gz
+ADNI10927885,F,57.1,adni_storage/ADNI_nii_gz_bias_corrected/I10927885_035_S_10193.stripped.N4.nii.gz
+sub-CC120120,M,25.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120120.stripped.N4.nii.gz
+fcon1000NewYork_sub20676,M,49.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub20676.stripped.N4.nii.gz
+fcon1000Milwaukee_sub67948,M,48.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub67948.stripped.N4.nii.gz
+ADNI1038250,F,60.3,adni_storage/ADNI_nii_gz_bias_corrected/I1038250_067_S_6529.stripped.N4.nii.gz
+sub-CC420582,M,54.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420582.stripped.N4.nii.gz
+fcon1000Cambridge_sub47498,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub47498.stripped.N4.nii.gz
+DLBS0028427_session_1,M,25.7,dlbs_storage/DLBS_bias_corrected/0028427_session_1.stripped.N4.nii.gz
+DLBS0028328_session_1,M,30.47,dlbs_storage/DLBS_bias_corrected/0028328_session_1.stripped.N4.nii.gz
+ADNI10883231,F,56.4,adni_storage/ADNI_nii_gz_bias_corrected/I10883231_941_S_10169.stripped.N4.nii.gz
+fcon1000Bangor_sub66585,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub66585.stripped.N4.nii.gz
+fcon1000NewYork_sub22608,F,23.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub22608.stripped.N4.nii.gz
+fcon1000Berlin_sub12855,M,33.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub12855.stripped.N4.nii.gz
+DLBS0028436_session_1,F,75.99,dlbs_storage/DLBS_bias_corrected/0028436_session_1.stripped.N4.nii.gz
+sub-CC420217,M,50.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420217.stripped.N4.nii.gz
+ADNI1221673,F,63.4,adni_storage/ADNI_nii_gz_bias_corrected/I1221673_013_S_6780.stripped.N4.nii.gz
+sub-CC721504,M,82.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721504.stripped.N4.nii.gz
+sub-CC620490,F,74.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620490.stripped.N4.nii.gz
+fcon1000Milwaukee_sub44912,F,49.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub44912.stripped.N4.nii.gz
+fcon1000Milwaukee_sub56084,M,53.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub56084.stripped.N4.nii.gz
+fcon1000Newark_sub49705,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub49705.stripped.N4.nii.gz
+sub-CC410179,M,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410179.stripped.N4.nii.gz
+DLBS0028460_session_1,F,37.05,dlbs_storage/DLBS_bias_corrected/0028460_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub92288,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub92288.stripped.N4.nii.gz
+ADNI308627,M,57.0,adni_storage/ADNI_nii_gz_bias_corrected/I308627_014_S_4039.stripped.N4.nii.gz
+fcon1000NewYork_sub16607,M,21.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub16607.stripped.N4.nii.gz
+sub-CC711027,M,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC711027.stripped.N4.nii.gz
+ADNI1003993,F,64.1,adni_storage/ADNI_nii_gz_bias_corrected/I1003993_023_S_6399.stripped.N4.nii.gz
+sub-CC420226,M,50.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420226.stripped.N4.nii.gz
+DLBS0028458_session_1,F,35.43,dlbs_storage/DLBS_bias_corrected/0028458_session_1.stripped.N4.nii.gz
+fcon1000Bangor_sub91556,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub91556.stripped.N4.nii.gz
+ADNI99327,M,61.1,adni_storage/ADNI_nii_gz_bias_corrected/I99327_020_S_1288.stripped.N4.nii.gz
+fcon1000Cambridge_sub58360,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub58360.stripped.N4.nii.gz
+ADNI238745,M,56.0,adni_storage/ADNI_nii_gz_bias_corrected/I238745_014_S_4039.stripped.N4.nii.gz
+fcon1000Cambridge_sub27613,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub27613.stripped.N4.nii.gz
+DLBS0028547_session_1,F,32.13,dlbs_storage/DLBS_bias_corrected/0028547_session_1.stripped.N4.nii.gz
+fcon1000NewYork_sub17109,M,46.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub17109.stripped.N4.nii.gz
+sub-CC420004,F,49.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420004.stripped.N4.nii.gz
+sub-CC720071,M,82.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720071.stripped.N4.nii.gz
+sub-CC721585,F,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721585.stripped.N4.nii.gz
+fcon1000Cambridge_sub10619,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub10619.stripped.N4.nii.gz
+DLBS0028555_session_1,M,64.86,dlbs_storage/DLBS_bias_corrected/0028555_session_1.stripped.N4.nii.gz
+ADNI10968674,F,59.2,adni_storage/ADNI_nii_gz_bias_corrected/I10968674_052_S_10252.stripped.N4.nii.gz
+sub-CC620406,M,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620406.stripped.N4.nii.gz
+fcon1000Beijing_sub10186,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub10186.stripped.N4.nii.gz
+fcon1000Atlanta_sub49816,M,27.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub49816.stripped.N4.nii.gz
+sub-CC520424,M,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520424.stripped.N4.nii.gz
+fcon1000Berlin_sub38279,M,29.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub38279.stripped.N4.nii.gz
+ADNI10277636,F,63.6,adni_storage/ADNI_nii_gz_bias_corrected/I10277636_033_S_10005.stripped.N4.nii.gz
+fcon1000Beijing_sub34943,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub34943.stripped.N4.nii.gz
+fcon1000Cambridge_sub90674,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub90674.stripped.N4.nii.gz
+fcon1000Beijing_sub85818,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub85818.stripped.N4.nii.gz
+DLBS0028357_session_1,F,56.36,dlbs_storage/DLBS_bias_corrected/0028357_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub23869,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub23869.stripped.N4.nii.gz
+ADNI280812,F,63.9,adni_storage/ADNI_nii_gz_bias_corrected/I280812_023_S_4448.stripped.N4.nii.gz
+sub-CC520097,F,62.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520097.stripped.N4.nii.gz
+sub-CC620085,F,70.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620085.stripped.N4.nii.gz
+DLBS0028570_session_1,F,63.31,dlbs_storage/DLBS_bias_corrected/0028570_session_1.stripped.N4.nii.gz
+sub-CC610146,F,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610146.stripped.N4.nii.gz
+sub-CC620919,M,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620919.stripped.N4.nii.gz
+sub-CC720941,M,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720941.stripped.N4.nii.gz
+ADNI423071,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I423071_128_S_4607.stripped.N4.nii.gz
+fcon1000Beijing_sub76377,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub76377.stripped.N4.nii.gz
+DLBS0028444_session_1,F,81.41,dlbs_storage/DLBS_bias_corrected/0028444_session_1.stripped.N4.nii.gz
+ADNI290867,M,60.0,adni_storage/ADNI_nii_gz_bias_corrected/I290867_128_S_4599.stripped.N4.nii.gz
+fcon1000Baltimore_sub54329,F,30.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub54329.stripped.N4.nii.gz
+ADNI998806,F,62.0,adni_storage/ADNI_nii_gz_bias_corrected/I998806_037_S_6271.stripped.N4.nii.gz
+ADNI285070,M,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I285070_073_S_4552.stripped.N4.nii.gz
+sub-CC520083,F,65.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520083.stripped.N4.nii.gz
+DLBS0028552_session_1,F,48.86,dlbs_storage/DLBS_bias_corrected/0028552_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub42253,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub42253.stripped.N4.nii.gz
+fcon1000Beijing_sub75878,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub75878.stripped.N4.nii.gz
+ADNI1611592,M,58.2,adni_storage/ADNI_nii_gz_bias_corrected/I1611592_033_S_6976.stripped.N4.nii.gz
+ADNI352948,M,64.0,adni_storage/ADNI_nii_gz_bias_corrected/I352948_002_S_1155.stripped.N4.nii.gz
+fcon1000Cambridge_sub35430,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub35430.stripped.N4.nii.gz
+fcon1000Cambridge_sub23780,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub23780.stripped.N4.nii.gz
+sub-CC110182,F,18.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110182.stripped.N4.nii.gz
+fcon1000Atlanta_sub72971,M,30.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub72971.stripped.N4.nii.gz
+ADNI10283161,M,59.2,adni_storage/ADNI_nii_gz_bias_corrected/I10283161_941_S_6998.stripped.N4.nii.gz
+sub-CC420089,F,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420089.stripped.N4.nii.gz
+sub-CC711244,F,85.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC711244.stripped.N4.nii.gz
+fcon1000Cleveland_sub61868,M,48.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub61868.stripped.N4.nii.gz
+ADNI1184048,F,62.6,adni_storage/ADNI_nii_gz_bias_corrected/I1184048_022_S_6013.stripped.N4.nii.gz
+ADNI966268,F,62.1,adni_storage/ADNI_nii_gz_bias_corrected/I966268_020_S_6227.stripped.N4.nii.gz
+fcon1000Cambridge_sub62908,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub62908.stripped.N4.nii.gz
+fcon1000Beijing_sub96163,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub96163.stripped.N4.nii.gz
+fcon1000Munchen_sub36052,F,63.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub36052.stripped.N4.nii.gz
+ADNI1067189,M,64.1,adni_storage/ADNI_nii_gz_bias_corrected/I1067189_012_S_6073.stripped.N4.nii.gz
+DLBS0028424_session_1,F,77.79,dlbs_storage/DLBS_bias_corrected/0028424_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub04665,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub04665.stripped.N4.nii.gz
+fcon1000Berlin_sub06716,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub06716.stripped.N4.nii.gz
+ADNI10883247,M,60.0,adni_storage/ADNI_nii_gz_bias_corrected/I10883247_037_S_10117.stripped.N4.nii.gz
+sub-CC322186,M,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC322186.stripped.N4.nii.gz
+sub-CC410243,F,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410243.stripped.N4.nii.gz
+DLBS0028336_session_1,F,80.65,dlbs_storage/DLBS_bias_corrected/0028336_session_1.stripped.N4.nii.gz
+ADNI263858,F,56.5,adni_storage/ADNI_nii_gz_bias_corrected/I263858_031_S_4024.stripped.N4.nii.gz
+sub-CC320616,M,39.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320616.stripped.N4.nii.gz
+fcon1000Cambridge_sub27230,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub27230.stripped.N4.nii.gz
+sub-CC120166,M,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120166.stripped.N4.nii.gz
+sub-CC610658,M,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610658.stripped.N4.nii.gz
+ADNI1608103,M,63.4,adni_storage/ADNI_nii_gz_bias_corrected/I1608103_016_S_6941.stripped.N4.nii.gz
+ADNI70745,M,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I70745_016_S_0769.stripped.N4.nii.gz
+ADNI10546533,F,58.6,adni_storage/ADNI_nii_gz_bias_corrected/I10546533_137_S_6906.stripped.N4.nii.gz
+fcon1000Beijing_sub07716,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub07716.stripped.N4.nii.gz
+fcon1000Beijing_sub33943,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub33943.stripped.N4.nii.gz
+sub-CC720329,M,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720329.stripped.N4.nii.gz
+ADNI1488791,F,54.9,adni_storage/ADNI_nii_gz_bias_corrected/I1488791_037_S_6951.stripped.N4.nii.gz
+ADNI270530,M,56.5,adni_storage/ADNI_nii_gz_bias_corrected/I270530_014_S_4039.stripped.N4.nii.gz
+fcon1000Cambridge_sub52442,F,25.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub52442.stripped.N4.nii.gz
+sub-CC510392,M,67.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510392.stripped.N4.nii.gz
+sub-CC222120,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222120.stripped.N4.nii.gz
+fcon1000Atlanta_sub18219,F,33.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub18219.stripped.N4.nii.gz
+ADNI417668,M,62.0,adni_storage/ADNI_nii_gz_bias_corrected/I417668_128_S_4599.stripped.N4.nii.gz
+sub-CC520745,F,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520745.stripped.N4.nii.gz
+ADNI879209,M,56.0,adni_storage/ADNI_nii_gz_bias_corrected/I879209_114_S_6039.stripped.N4.nii.gz
+fcon1000Beijing_sub36580,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub36580.stripped.N4.nii.gz
+fcon1000Cambridge_sub50953,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub50953.stripped.N4.nii.gz
+fcon1000AnnArbor_sub62819,M,65.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub62819.stripped.N4.nii.gz
+sub-CC222797,F,31.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222797.stripped.N4.nii.gz
+DLBS0028546_session_1,M,21.71,dlbs_storage/DLBS_bias_corrected/0028546_session_1.stripped.N4.nii.gz
+sub-CC112141,M,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC112141.stripped.N4.nii.gz
+fcon1000Beijing_sub73279,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub73279.stripped.N4.nii.gz
+sub-CC621080,M,70.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC621080.stripped.N4.nii.gz
+fcon1000Cambridge_sub77435,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub77435.stripped.N4.nii.gz
+sub-CC310402,F,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310402.stripped.N4.nii.gz
+fcon1000Beijing_sub28206,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub28206.stripped.N4.nii.gz
+fcon1000Munchen_sub28697,F,73.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub28697.stripped.N4.nii.gz
+DLBS0028487_session_1,F,73.56,dlbs_storage/DLBS_bias_corrected/0028487_session_1.stripped.N4.nii.gz
+sub-CC710566,M,83.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710566.stripped.N4.nii.gz
+fcon1000Cambridge_sub93609,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub93609.stripped.N4.nii.gz
+sub-CC121158,F,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121158.stripped.N4.nii.gz
+sub-CC620279,M,77.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620279.stripped.N4.nii.gz
+fcon1000Baltimore_sub23750,M,37.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub23750.stripped.N4.nii.gz
+sub-CC310142,M,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310142.stripped.N4.nii.gz
+ADNI442428,F,59.8,adni_storage/ADNI_nii_gz_bias_corrected/I442428_041_S_4874.stripped.N4.nii.gz
+DLBS0028337_session_1,F,53.34,dlbs_storage/DLBS_bias_corrected/0028337_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub11072,F,25.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub11072.stripped.N4.nii.gz
+ADNI10461519,F,61.9,adni_storage/ADNI_nii_gz_bias_corrected/I10461519_022_S_10043.stripped.N4.nii.gz
+fcon1000NewYork_sub20691,M,26.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub20691.stripped.N4.nii.gz
+fcon1000Beijing_sub48501,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub48501.stripped.N4.nii.gz
+sub-CC220223,M,33.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220223.stripped.N4.nii.gz
+ADNI907713,F,58.7,adni_storage/ADNI_nii_gz_bias_corrected/I907713_014_S_6076.stripped.N4.nii.gz
+fcon1000AnnArbor_sub51248,M,40.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub51248.stripped.N4.nii.gz
+DLBS0028483_session_1,F,47.0,dlbs_storage/DLBS_bias_corrected/0028483_session_1.stripped.N4.nii.gz
+fcon1000Newark_sub59397,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub59397.stripped.N4.nii.gz
+fcon1000Beijing_sub73098,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub73098.stripped.N4.nii.gz
+sub-CC420566,F,50.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420566.stripped.N4.nii.gz
+fcon1000Berlin_sub91116,M,39.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub91116.stripped.N4.nii.gz
+sub-CC620466,M,75.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620466.stripped.N4.nii.gz
+sub-CC310463,F,45.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310463.stripped.N4.nii.gz
+ADNI1438743,F,64.3,adni_storage/ADNI_nii_gz_bias_corrected/I1438743_114_S_6917.stripped.N4.nii.gz
+DLBS0028389_session_1,M,33.45,dlbs_storage/DLBS_bias_corrected/0028389_session_1.stripped.N4.nii.gz
+ADNI386092,F,58.8,adni_storage/ADNI_nii_gz_bias_corrected/I386092_041_S_4874.stripped.N4.nii.gz
+ADNI341369,M,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I341369_116_S_4625.stripped.N4.nii.gz
+sub-CC721107,F,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721107.stripped.N4.nii.gz
+fcon1000Cambridge_sub21755,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub21755.stripped.N4.nii.gz
+sub-CC520211,F,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520211.stripped.N4.nii.gz
+sub-CC220999,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220999.stripped.N4.nii.gz
+sub-CC710446,M,85.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710446.stripped.N4.nii.gz
+fcon1000Beijing_sub82714,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub82714.stripped.N4.nii.gz
+ADNI296876,F,58.5,adni_storage/ADNI_nii_gz_bias_corrected/I296876_012_S_4642.stripped.N4.nii.gz
+fcon1000Beijing_sub30556,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub30556.stripped.N4.nii.gz
+fcon1000Beijing_sub28801,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub28801.stripped.N4.nii.gz
+ADNI86563,F,57.3,adni_storage/ADNI_nii_gz_bias_corrected/I86563_016_S_1121.stripped.N4.nii.gz
+fcon1000Beijing_sub17688,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub17688.stripped.N4.nii.gz
+sub-CC610178,M,72.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610178.stripped.N4.nii.gz
+ADNI1577482,F,56.7,adni_storage/ADNI_nii_gz_bias_corrected/I1577482_305_S_6899.stripped.N4.nii.gz
+DLBS0028475_session_1,M,54.05,dlbs_storage/DLBS_bias_corrected/0028475_session_1.stripped.N4.nii.gz
+ADNI271073,M,62.6,adni_storage/ADNI_nii_gz_bias_corrected/I271073_116_S_4195.stripped.N4.nii.gz
+DLBS0028531_session_1,F,21.27,dlbs_storage/DLBS_bias_corrected/0028531_session_1.stripped.N4.nii.gz
+sub-CC120550,M,19.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120550.stripped.N4.nii.gz
+fcon1000Cambridge_sub73399,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub73399.stripped.N4.nii.gz
+fcon1000Beijing_sub86114,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub86114.stripped.N4.nii.gz
+fcon1000Beijing_sub62966,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub62966.stripped.N4.nii.gz
+DLBS0028384_session_1,F,30.5,dlbs_storage/DLBS_bias_corrected/0028384_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub66781,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub66781.stripped.N4.nii.gz
+sub-CC221733,M,36.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221733.stripped.N4.nii.gz
+fcon1000Beijing_sub66158,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub66158.stripped.N4.nii.gz
+sub-CC610344,F,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610344.stripped.N4.nii.gz
+fcon1000Milwaukee_sub49975,M,46.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub49975.stripped.N4.nii.gz
+DLBS0028332_session_1,M,71.29,dlbs_storage/DLBS_bias_corrected/0028332_session_1.stripped.N4.nii.gz
+DLBS0028471_session_1,F,67.47,dlbs_storage/DLBS_bias_corrected/0028471_session_1.stripped.N4.nii.gz
+ADNI345557,F,63.0,adni_storage/ADNI_nii_gz_bias_corrected/I345557_019_S_5019.stripped.N4.nii.gz
+fcon1000Leiden_sub52922,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub52922.stripped.N4.nii.gz
+sub-CC210250,M,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210250.stripped.N4.nii.gz
+ADNI1092245,M,57.4,adni_storage/ADNI_nii_gz_bias_corrected/I1092245_114_S_6039.stripped.N4.nii.gz
+fcon1000Newark_sub43517,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub43517.stripped.N4.nii.gz
+fcon1000Bangor_sub00031,M,29.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub00031.stripped.N4.nii.gz
+fcon1000NewYork_sub46856,F,8.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub46856.stripped.N4.nii.gz
+ADNI1032324,F,56.5,adni_storage/ADNI_nii_gz_bias_corrected/I1032324_070_S_6548.stripped.N4.nii.gz
+sub-CC610594,F,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610594.stripped.N4.nii.gz
+fcon1000Munchen_sub09035,M,70.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub09035.stripped.N4.nii.gz
+fcon1000Beijing_sub65659,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub65659.stripped.N4.nii.gz
+fcon1000Cambridge_sub05453,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub05453.stripped.N4.nii.gz
+fcon1000Cambridge_sub51050,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub51050.stripped.N4.nii.gz
+sub-CC410173,M,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410173.stripped.N4.nii.gz
+DLBS0028407_session_1,M,86.86,dlbs_storage/DLBS_bias_corrected/0028407_session_1.stripped.N4.nii.gz
+ADNI1304804,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I1304804_168_S_6085.stripped.N4.nii.gz
+ADNI365456,M,62.1,adni_storage/ADNI_nii_gz_bias_corrected/I365456_073_S_4795.stripped.N4.nii.gz
+fcon1000Milwaukee_sub98971,F,63.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub98971.stripped.N4.nii.gz
+sub-CC420261,F,54.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420261.stripped.N4.nii.gz
+DLBS0028390_session_1,M,22.5,dlbs_storage/DLBS_bias_corrected/0028390_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub73317,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub73317.stripped.N4.nii.gz
+fcon1000Cambridge_sub67117,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub67117.stripped.N4.nii.gz
+ADNI1457865,F,64.5,adni_storage/ADNI_nii_gz_bias_corrected/I1457865_137_S_6903.stripped.N4.nii.gz
+DLBS0028351_session_1,M,83.73,dlbs_storage/DLBS_bias_corrected/0028351_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub50859,F,79.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub50859.stripped.N4.nii.gz
+ADNI1173877,F,62.2,adni_storage/ADNI_nii_gz_bias_corrected/I1173877_067_S_6525.stripped.N4.nii.gz
+fcon1000Beijing_sub92430,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub92430.stripped.N4.nii.gz
+ADNI952046,F,60.2,adni_storage/ADNI_nii_gz_bias_corrected/I952046_037_S_6141.stripped.N4.nii.gz
+fcon1000Cambridge_sub88853,F,29.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub88853.stripped.N4.nii.gz
+DLBS0028575_session_1,M,81.02,dlbs_storage/DLBS_bias_corrected/0028575_session_1.stripped.N4.nii.gz
+sub-CC220713,F,33.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220713.stripped.N4.nii.gz
+DLBS0028600_session_1,M,70.31,dlbs_storage/DLBS_bias_corrected/0028600_session_1.stripped.N4.nii.gz
+sub-CC110174,F,25.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110174.stripped.N4.nii.gz
+fcon1000Beijing_sub59448,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub59448.stripped.N4.nii.gz
+fcon1000Berlin_sub95068,M,26.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub95068.stripped.N4.nii.gz
+ADNI944155,F,62.2,adni_storage/ADNI_nii_gz_bias_corrected/I944155_116_S_6129.stripped.N4.nii.gz
+sub-CC220419,M,31.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220419.stripped.N4.nii.gz
+fcon1000Beijing_sub17642,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub17642.stripped.N4.nii.gz
+sub-CC520215,F,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520215.stripped.N4.nii.gz
+fcon1000Bangor_sub36736,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub36736.stripped.N4.nii.gz
+DLBS0028480_session_1,F,40.26,dlbs_storage/DLBS_bias_corrected/0028480_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub13093,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub13093.stripped.N4.nii.gz
+fcon1000NewYork_sub69779,M,40.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub69779.stripped.N4.nii.gz
+fcon1000Cambridge_sub12346,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub12346.stripped.N4.nii.gz
+ADNI62032,F,56.8,adni_storage/ADNI_nii_gz_bias_corrected/I62032_016_S_1121.stripped.N4.nii.gz
+sub-CC410169,F,49.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410169.stripped.N4.nii.gz
+sub-CC310450,F,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310450.stripped.N4.nii.gz
+DLBS0028520_session_1,F,65.52,dlbs_storage/DLBS_bias_corrected/0028520_session_1.stripped.N4.nii.gz
+sub-CC510639,M,62.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510639.stripped.N4.nii.gz
+sub-CC420286,M,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420286.stripped.N4.nii.gz
+fcon1000Leiden_sub57187,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub57187.stripped.N4.nii.gz
+sub-CC312222,F,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC312222.stripped.N4.nii.gz
+fcon1000Bangor_sub03557,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub03557.stripped.N4.nii.gz
+DLBS0028545_session_1,F,72.25,dlbs_storage/DLBS_bias_corrected/0028545_session_1.stripped.N4.nii.gz
+fcon1000Leiden_sub38454,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub38454.stripped.N4.nii.gz
+fcon1000Beijing_sub17603,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub17603.stripped.N4.nii.gz
+DLBS0028405_session_1,F,86.45,dlbs_storage/DLBS_bias_corrected/0028405_session_1.stripped.N4.nii.gz
+sub-CC220284,F,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220284.stripped.N4.nii.gz
+DLBS0028629_session_1,F,81.12,dlbs_storage/DLBS_bias_corrected/0028629_session_1.stripped.N4.nii.gz
+sub-CC620454,F,75.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620454.stripped.N4.nii.gz
+ADNI10952529,M,63.9,adni_storage/ADNI_nii_gz_bias_corrected/I10952529_037_S_10062.stripped.N4.nii.gz
+sub-CC420776,F,49.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420776.stripped.N4.nii.gz
+fcon1000Cambridge_sub55114,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub55114.stripped.N4.nii.gz
+sub-CC210526,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210526.stripped.N4.nii.gz
+fcon1000AnnArbor_sub04111,M,25.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub04111.stripped.N4.nii.gz
+fcon1000Beijing_sub45552,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub45552.stripped.N4.nii.gz
+sub-CC710214,M,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710214.stripped.N4.nii.gz
+sub-CC510329,M,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510329.stripped.N4.nii.gz
+DLBS0028454_session_1,M,42.92,dlbs_storage/DLBS_bias_corrected/0028454_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub30421,M,17.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub30421.stripped.N4.nii.gz
+DLBS0028368_session_1,M,79.65,dlbs_storage/DLBS_bias_corrected/0028368_session_1.stripped.N4.nii.gz
+fcon1000NewYork_sub84371,M,38.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub84371.stripped.N4.nii.gz
+ADNI1558064,F,61.8,adni_storage/ADNI_nii_gz_bias_corrected/I1558064_011_S_7048.stripped.N4.nii.gz
+fcon1000Cambridge_sub00156,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub00156.stripped.N4.nii.gz
+sub-CC510255,M,62.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510255.stripped.N4.nii.gz
+sub-CC210304,F,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210304.stripped.N4.nii.gz
+fcon1000NewYork_sub29935,F,16.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub29935.stripped.N4.nii.gz
+fcon1000Beijing_sub73421,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub73421.stripped.N4.nii.gz
+DLBS0028369_session_1,F,62.67,dlbs_storage/DLBS_bias_corrected/0028369_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub20317,M,15.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub20317.stripped.N4.nii.gz
+sub-CC320870,F,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320870.stripped.N4.nii.gz
+fcon1000Cambridge_sub14194,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub14194.stripped.N4.nii.gz
+sub-CC520391,F,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520391.stripped.N4.nii.gz
+sub-CC110319,F,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110319.stripped.N4.nii.gz
+fcon1000AnnArbor_sub72416,M,80.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub72416.stripped.N4.nii.gz
+ADNI263857,F,56.5,adni_storage/ADNI_nii_gz_bias_corrected/I263857_031_S_4024.stripped.N4.nii.gz
+ADNI10888807,M,60.3,adni_storage/ADNI_nii_gz_bias_corrected/I10888807_019_S_10164.stripped.N4.nii.gz
+fcon1000AnnArbor_sub70106,M,17.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub70106.stripped.N4.nii.gz
+DLBS0028558_session_1,F,24.83,dlbs_storage/DLBS_bias_corrected/0028558_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub34781,M,19.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub34781.stripped.N4.nii.gz
+fcon1000Atlanta_sub32093,F,27.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub32093.stripped.N4.nii.gz
+sub-CC520560,M,65.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520560.stripped.N4.nii.gz
+sub-CC621248,F,72.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC621248.stripped.N4.nii.gz
+sub-CC420396,M,53.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420396.stripped.N4.nii.gz
+ADNI1146201,F,61.4,adni_storage/ADNI_nii_gz_bias_corrected/I1146201_037_S_6141.stripped.N4.nii.gz
+sub-CC122172,M,27.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC122172.stripped.N4.nii.gz
+fcon1000Beijing_sub11344,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub11344.stripped.N4.nii.gz
+sub-CC321000,M,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321000.stripped.N4.nii.gz
+sub-CC620026,F,70.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620026.stripped.N4.nii.gz
+sub-CC320500,M,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320500.stripped.N4.nii.gz
+sub-CC721052,M,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721052.stripped.N4.nii.gz
+fcon1000Beijing_sub28907,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub28907.stripped.N4.nii.gz
+fcon1000Cambridge_sub08204,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub08204.stripped.N4.nii.gz
+fcon1000Cambridge_sub16846,M,29.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub16846.stripped.N4.nii.gz
+DLBS0028341_session_1,F,34.81,dlbs_storage/DLBS_bias_corrected/0028341_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub52300,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub52300.stripped.N4.nii.gz
+DLBS0028393_session_1,M,61.69,dlbs_storage/DLBS_bias_corrected/0028393_session_1.stripped.N4.nii.gz
+DLBS0028421_session_1,M,67.05,dlbs_storage/DLBS_bias_corrected/0028421_session_1.stripped.N4.nii.gz
+sub-CC520013,M,67.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520013.stripped.N4.nii.gz
+sub-CC210314,F,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210314.stripped.N4.nii.gz
+fcon1000AnnArbor_sub86367,M,15.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub86367.stripped.N4.nii.gz
+sub-CC420143,F,53.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420143.stripped.N4.nii.gz
+sub-CC320206,M,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320206.stripped.N4.nii.gz
+DLBS0028492_session_1,M,52.89,dlbs_storage/DLBS_bias_corrected/0028492_session_1.stripped.N4.nii.gz
+sub-CC321506,F,39.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321506.stripped.N4.nii.gz
+DLBS0028493_session_1,M,53.36,dlbs_storage/DLBS_bias_corrected/0028493_session_1.stripped.N4.nii.gz
+ADNI1461223,M,61.6,adni_storage/ADNI_nii_gz_bias_corrected/I1461223_070_S_6966.stripped.N4.nii.gz
+fcon1000Beijing_sub30310,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub30310.stripped.N4.nii.gz
+ADNI1142218,F,60.9,adni_storage/ADNI_nii_gz_bias_corrected/I1142218_033_S_6697.stripped.N4.nii.gz
+sub-CC723197,F,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC723197.stripped.N4.nii.gz
+ADNI341787,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I341787_130_S_4982.stripped.N4.nii.gz
+DLBS0028333_session_1,F,76.45,dlbs_storage/DLBS_bias_corrected/0028333_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub95959,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub95959.stripped.N4.nii.gz
+DLBS0028530_session_1,F,49.86,dlbs_storage/DLBS_bias_corrected/0028530_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub22201,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub22201.stripped.N4.nii.gz
+ADNI321225,F,57.8,adni_storage/ADNI_nii_gz_bias_corrected/I321225_041_S_4874.stripped.N4.nii.gz
+ADNI1020137,F,62.6,adni_storage/ADNI_nii_gz_bias_corrected/I1020137_019_S_6483.stripped.N4.nii.gz
+ADNI1553005,M,63.7,adni_storage/ADNI_nii_gz_bias_corrected/I1553005_941_S_7051.stripped.N4.nii.gz
+ADNI297850,F,57.0,adni_storage/ADNI_nii_gz_bias_corrected/I297850_031_S_4024.stripped.N4.nii.gz
+ADNI1591455,F,62.7,adni_storage/ADNI_nii_gz_bias_corrected/I1591455_014_S_7072.stripped.N4.nii.gz
+DLBS0028468_session_1,F,44.31,dlbs_storage/DLBS_bias_corrected/0028468_session_1.stripped.N4.nii.gz
+ADNI400949,M,62.8,adni_storage/ADNI_nii_gz_bias_corrected/I400949_073_S_4795.stripped.N4.nii.gz
+sub-CC510259,F,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510259.stripped.N4.nii.gz
+ADNI1061955,F,58.0,adni_storage/ADNI_nii_gz_bias_corrected/I1061955_109_S_6364.stripped.N4.nii.gz
+fcon1000Cleveland_sub47482,F,28.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub47482.stripped.N4.nii.gz
+sub-CC620005,F,74.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620005.stripped.N4.nii.gz
+DLBS0028413_session_1,F,81.57,dlbs_storage/DLBS_bias_corrected/0028413_session_1.stripped.N4.nii.gz
+DLBS0028415_session_1,F,32.05,dlbs_storage/DLBS_bias_corrected/0028415_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub64969,F,21.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub64969.stripped.N4.nii.gz
+sub-CC510256,M,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510256.stripped.N4.nii.gz
+fcon1000Leiden_sub09796,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub09796.stripped.N4.nii.gz
+sub-CC310410,F,41.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310410.stripped.N4.nii.gz
+ADNI1565078,M,57.6,adni_storage/ADNI_nii_gz_bias_corrected/I1565078_067_S_7059.stripped.N4.nii.gz
+sub-CC310052,F,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310052.stripped.N4.nii.gz
+sub-CC420236,M,53.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420236.stripped.N4.nii.gz
+ADNI363013,F,61.4,adni_storage/ADNI_nii_gz_bias_corrected/I363013_036_S_4894.stripped.N4.nii.gz
+DLBS0028522_session_1,F,34.99,dlbs_storage/DLBS_bias_corrected/0028522_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub91399,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub91399.stripped.N4.nii.gz
+fcon1000NewYork_sub38088,M,16.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub38088.stripped.N4.nii.gz
+DLBS0028348_session_1,M,23.15,dlbs_storage/DLBS_bias_corrected/0028348_session_1.stripped.N4.nii.gz
+fcon1000NewYork_sub48830,F,22.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub48830.stripped.N4.nii.gz
+fcon1000AnnArbor_sub39635,F,19.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub39635.stripped.N4.nii.gz
+fcon1000NewYork_sub31671,M,12.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub31671.stripped.N4.nii.gz
+fcon1000Cambridge_sub76745,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub76745.stripped.N4.nii.gz
+fcon1000Beijing_sub89592,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub89592.stripped.N4.nii.gz
+ADNI387592,F,61.8,adni_storage/ADNI_nii_gz_bias_corrected/I387592_036_S_4894.stripped.N4.nii.gz
+ADNI247840,F,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I247840_037_S_4028.stripped.N4.nii.gz
+fcon1000Beijing_sub18326,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub18326.stripped.N4.nii.gz
+ADNI1634179,F,58.0,adni_storage/ADNI_nii_gz_bias_corrected/I1634179_013_S_7103.stripped.N4.nii.gz
+sub-CC710154,M,83.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710154.stripped.N4.nii.gz
+sub-CC410097,F,49.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410097.stripped.N4.nii.gz
+fcon1000Beijing_sub85543,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub85543.stripped.N4.nii.gz
+ADNI105503,F,63.9,adni_storage/ADNI_nii_gz_bias_corrected/I105503_067_S_1253.stripped.N4.nii.gz
+fcon1000AnnArbor_sub75922,M,18.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub75922.stripped.N4.nii.gz
+sub-CC220151,M,30.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220151.stripped.N4.nii.gz
+fcon1000Cambridge_sub68101,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub68101.stripped.N4.nii.gz
+sub-CC120347,F,21.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120347.stripped.N4.nii.gz
+fcon1000NewYork_sub73035,M,26.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub73035.stripped.N4.nii.gz
+sub-CC620262,F,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620262.stripped.N4.nii.gz
+DLBS0028383_session_1,M,72.96,dlbs_storage/DLBS_bias_corrected/0028383_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub53998,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub53998.stripped.N4.nii.gz
+fcon1000Beijing_sub92859,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub92859.stripped.N4.nii.gz
+ADNI1482729,F,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I1482729_067_S_6529.stripped.N4.nii.gz
+ADNI10951704,M,64.2,adni_storage/ADNI_nii_gz_bias_corrected/I10951704_035_S_10241.stripped.N4.nii.gz
+fcon1000Milwaukee_sub21350,F,61.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub21350.stripped.N4.nii.gz
+DLBS0028463_session_1,F,68.35,dlbs_storage/DLBS_bias_corrected/0028463_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub88464,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub88464.stripped.N4.nii.gz
+fcon1000Cambridge_sub78547,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub78547.stripped.N4.nii.gz
+sub-CC110411,M,25.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110411.stripped.N4.nii.gz
+sub-CC510242,M,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510242.stripped.N4.nii.gz
+ADNI1469825,F,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I1469825_073_S_6929.stripped.N4.nii.gz
+ADNI1028361,M,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I1028361_109_S_6406.stripped.N4.nii.gz
+ADNI346860,F,56.9,adni_storage/ADNI_nii_gz_bias_corrected/I346860_128_S_4607.stripped.N4.nii.gz
+fcon1000Cambridge_sub77989,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub77989.stripped.N4.nii.gz
+ADNI10955657,M,58.4,adni_storage/ADNI_nii_gz_bias_corrected/I10955657_389_S_10156.stripped.N4.nii.gz
+fcon1000Cambridge_sub15905,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub15905.stripped.N4.nii.gz
+sub-CC720103,M,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720103.stripped.N4.nii.gz
+fcon1000NewYork_sub07578,F,20.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub07578.stripped.N4.nii.gz
+sub-CC221648,F,30.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221648.stripped.N4.nii.gz
+ADNI47973,F,62.8,adni_storage/ADNI_nii_gz_bias_corrected/I47973_067_S_1253.stripped.N4.nii.gz
+DLBS0028539_session_1,F,24.22,dlbs_storage/DLBS_bias_corrected/0028539_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub20948,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub20948.stripped.N4.nii.gz
+fcon1000Baltimore_sub37548,M,30.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub37548.stripped.N4.nii.gz
+fcon1000Beijing_sub31729,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub31729.stripped.N4.nii.gz
+sub-CC221755,M,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221755.stripped.N4.nii.gz
+sub-CC620685,F,77.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620685.stripped.N4.nii.gz
+sub-CC121200,F,26.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121200.stripped.N4.nii.gz
+fcon1000Newark_sub32580,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub32580.stripped.N4.nii.gz
+DLBS0028540_session_1,F,71.04,dlbs_storage/DLBS_bias_corrected/0028540_session_1.stripped.N4.nii.gz
+sub-CC110101,M,23.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110101.stripped.N4.nii.gz
+fcon1000NewYork_sub50559,M,22.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub50559.stripped.N4.nii.gz
+DLBS0028589_session_1,F,59.75,dlbs_storage/DLBS_bias_corrected/0028589_session_1.stripped.N4.nii.gz
+fcon1000Baltimore_sub73823,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub73823.stripped.N4.nii.gz
+sub-CC321976,F,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321976.stripped.N4.nii.gz
+fcon1000Munchen_sub72095,F,71.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub72095.stripped.N4.nii.gz
+sub-CC220198,F,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220198.stripped.N4.nii.gz
+ADNI389099,F,59.6,adni_storage/ADNI_nii_gz_bias_corrected/I389099_067_S_5205.stripped.N4.nii.gz
+fcon1000Bangor_sub81464,M,38.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub81464.stripped.N4.nii.gz
+sub-CC221585,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221585.stripped.N4.nii.gz
+fcon1000Cambridge_sub45354,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub45354.stripped.N4.nii.gz
+fcon1000Beijing_sub16943,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub16943.stripped.N4.nii.gz
+ADNI10302840,F,59.5,adni_storage/ADNI_nii_gz_bias_corrected/I10302840_016_S_6773.stripped.N4.nii.gz
+sub-CC620090,F,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620090.stripped.N4.nii.gz
+fcon1000Bangor_sub82625,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub82625.stripped.N4.nii.gz
+ADNI1609469,F,60.8,adni_storage/ADNI_nii_gz_bias_corrected/I1609469_035_S_6927.stripped.N4.nii.gz
+sub-CC320002,M,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320002.stripped.N4.nii.gz
+fcon1000Milwaukee_sub87910,M,47.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub87910.stripped.N4.nii.gz
+DLBS0028499_session_1,M,51.56,dlbs_storage/DLBS_bias_corrected/0028499_session_1.stripped.N4.nii.gz
+sub-CC420202,M,51.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420202.stripped.N4.nii.gz
+fcon1000Cambridge_sub63412,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub63412.stripped.N4.nii.gz
+sub-CC510354,F,62.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510354.stripped.N4.nii.gz
+ADNI10244463,F,61.8,adni_storage/ADNI_nii_gz_bias_corrected/I10244463_035_S_6927.stripped.N4.nii.gz
+sub-CC510551,M,61.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510551.stripped.N4.nii.gz
+fcon1000Leiden_sub99856,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub99856.stripped.N4.nii.gz
+fcon1000AnnArbor_sub11043,F,21.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub11043.stripped.N4.nii.gz
+DLBS0028417_session_1,F,31.77,dlbs_storage/DLBS_bias_corrected/0028417_session_1.stripped.N4.nii.gz
+sub-CC121411,F,26.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121411.stripped.N4.nii.gz
+fcon1000Beijing_sub56136,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub56136.stripped.N4.nii.gz
+DLBS0028435_session_1,M,33.01,dlbs_storage/DLBS_bias_corrected/0028435_session_1.stripped.N4.nii.gz
+DLBS0028366_session_1,F,23.53,dlbs_storage/DLBS_bias_corrected/0028366_session_1.stripped.N4.nii.gz
+ADNI1590221,M,61.4,adni_storage/ADNI_nii_gz_bias_corrected/I1590221_053_S_7086.stripped.N4.nii.gz
+fcon1000Cambridge_sub58470,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub58470.stripped.N4.nii.gz
+fcon1000Milwaukee_sub00917,F,54.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub00917.stripped.N4.nii.gz
+ADNI290869,M,60.0,adni_storage/ADNI_nii_gz_bias_corrected/I290869_128_S_4599.stripped.N4.nii.gz
+ADNI140027,M,62.1,adni_storage/ADNI_nii_gz_bias_corrected/I140027_020_S_1288.stripped.N4.nii.gz
+fcon1000Beijing_sub50498,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub50498.stripped.N4.nii.gz
+DLBS0028380_session_1,F,83.22,dlbs_storage/DLBS_bias_corrected/0028380_session_1.stripped.N4.nii.gz
+DLBS0028636_session_1,F,65.12,dlbs_storage/DLBS_bias_corrected/0028636_session_1.stripped.N4.nii.gz
+fcon1000Munchen_sub70942,M,70.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub70942.stripped.N4.nii.gz
+ADNI1592054,F,57.3,adni_storage/ADNI_nii_gz_bias_corrected/I1592054_014_S_7080.stripped.N4.nii.gz
+DLBS0028378_session_1,F,69.19,dlbs_storage/DLBS_bias_corrected/0028378_session_1.stripped.N4.nii.gz
+fcon1000Leiden_sub12255,M,25.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub12255.stripped.N4.nii.gz
+sub-CC620121,M,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620121.stripped.N4.nii.gz
+sub-CC321087,F,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321087.stripped.N4.nii.gz
+ADNI1539706,F,59.7,adni_storage/ADNI_nii_gz_bias_corrected/I1539706_019_S_6635.stripped.N4.nii.gz
+ADNI73380,F,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I73380_051_S_1072.stripped.N4.nii.gz
+fcon1000Beijing_sub41621,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub41621.stripped.N4.nii.gz
+fcon1000Cambridge_sub78614,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub78614.stripped.N4.nii.gz
+ADNI342681,F,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I342681_073_S_5016.stripped.N4.nii.gz
+fcon1000Beijing_sub82826,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub82826.stripped.N4.nii.gz
+fcon1000Beijing_sub93856,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub93856.stripped.N4.nii.gz
+fcon1000Milwaukee_sub28782,F,64.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub28782.stripped.N4.nii.gz
+sub-CC410101,M,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410101.stripped.N4.nii.gz
+ADNI1583750,F,63.0,adni_storage/ADNI_nii_gz_bias_corrected/I1583750_067_S_7068.stripped.N4.nii.gz
+sub-CC610292,F,72.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610292.stripped.N4.nii.gz
+DLBS0028371_session_1,M,31.11,dlbs_storage/DLBS_bias_corrected/0028371_session_1.stripped.N4.nii.gz
+sub-CC120061,M,19.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120061.stripped.N4.nii.gz
+ADNI10454329,F,56.8,adni_storage/ADNI_nii_gz_bias_corrected/I10454329_003_S_10073.stripped.N4.nii.gz
+fcon1000Cambridge_sub40635,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub40635.stripped.N4.nii.gz
+DLBS0028581_session_1,F,74.58,dlbs_storage/DLBS_bias_corrected/0028581_session_1.stripped.N4.nii.gz
+ADNI302014,F,64.6,adni_storage/ADNI_nii_gz_bias_corrected/I302014_037_S_4028.stripped.N4.nii.gz
+sub-CC321544,M,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321544.stripped.N4.nii.gz
+fcon1000Beijing_sub08251,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub08251.stripped.N4.nii.gz
+DLBS0028448_session_1,M,69.32,dlbs_storage/DLBS_bias_corrected/0028448_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub83430,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub83430.stripped.N4.nii.gz
+fcon1000Cambridge_sub82113,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub82113.stripped.N4.nii.gz
+sub-CC410121,F,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410121.stripped.N4.nii.gz
+DLBS0028498_session_1,M,67.16,dlbs_storage/DLBS_bias_corrected/0028498_session_1.stripped.N4.nii.gz
+sub-CC520597,M,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520597.stripped.N4.nii.gz
+sub-CC420356,M,54.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420356.stripped.N4.nii.gz
+ADNI293371,F,56.3,adni_storage/ADNI_nii_gz_bias_corrected/I293371_128_S_4607.stripped.N4.nii.gz
+DLBS0028339_session_1,F,65.23,dlbs_storage/DLBS_bias_corrected/0028339_session_1.stripped.N4.nii.gz
+sub-CC420729,F,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420729.stripped.N4.nii.gz
+ADNI10369709,F,58.5,adni_storage/ADNI_nii_gz_bias_corrected/I10369709_073_S_10054.stripped.N4.nii.gz
+fcon1000Berlin_sub86111,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub86111.stripped.N4.nii.gz
+ADNI495308,F,57.3,adni_storage/ADNI_nii_gz_bias_corrected/I495308_128_S_4607.stripped.N4.nii.gz
+ADNI321230,F,57.8,adni_storage/ADNI_nii_gz_bias_corrected/I321230_041_S_4874.stripped.N4.nii.gz
+ADNI461101,F,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I461101_137_S_0994.stripped.N4.nii.gz
+DLBS0028403_session_1,M,68.71,dlbs_storage/DLBS_bias_corrected/0028403_session_1.stripped.N4.nii.gz
+fcon1000Leiden_sub30943,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub30943.stripped.N4.nii.gz
+fcon1000Baltimore_sub94042,F,27.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub94042.stripped.N4.nii.gz
+fcon1000Beijing_sub40427,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub40427.stripped.N4.nii.gz
+sub-CC721392,M,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721392.stripped.N4.nii.gz
+sub-CC520552,M,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520552.stripped.N4.nii.gz
+ADNI256321,M,56.3,adni_storage/ADNI_nii_gz_bias_corrected/I256321_014_S_4039.stripped.N4.nii.gz
+ADNI10568658,M,56.7,adni_storage/ADNI_nii_gz_bias_corrected/I10568658_941_S_10065.stripped.N4.nii.gz
+fcon1000Leiden_sub36743,M,25.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub36743.stripped.N4.nii.gz
+ADNI1498335,F,57.3,adni_storage/ADNI_nii_gz_bias_corrected/I1498335_016_S_6773.stripped.N4.nii.gz
+fcon1000AnnArbor_sub13636,M,29.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub13636.stripped.N4.nii.gz
+fcon1000AnnArbor_sub49687,M,13.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub49687.stripped.N4.nii.gz
+sub-CC320814,F,43.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320814.stripped.N4.nii.gz
+ADNI1446680,F,59.8,adni_storage/ADNI_nii_gz_bias_corrected/I1446680_070_S_6942.stripped.N4.nii.gz
+sub-CC620044,F,73.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620044.stripped.N4.nii.gz
+sub-CC510220,M,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510220.stripped.N4.nii.gz
+ADNI313456,M,60.3,adni_storage/ADNI_nii_gz_bias_corrected/I313456_128_S_4599.stripped.N4.nii.gz
+ADNI10970157,M,56.0,adni_storage/ADNI_nii_gz_bias_corrected/I10970157_073_S_10254.stripped.N4.nii.gz
+sub-CC412021,F,54.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC412021.stripped.N4.nii.gz
+sub-CC610212,F,77.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610212.stripped.N4.nii.gz
+sub-CC220843,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220843.stripped.N4.nii.gz
+sub-CC620610,M,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620610.stripped.N4.nii.gz
+ADNI1558652,F,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I1558652_014_S_6076.stripped.N4.nii.gz
+fcon1000Cambridge_sub65682,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub65682.stripped.N4.nii.gz
+DLBS0028385_session_1,F,61.53,dlbs_storage/DLBS_bias_corrected/0028385_session_1.stripped.N4.nii.gz
+sub-CC410432,M,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410432.stripped.N4.nii.gz
+sub-CC321529,F,39.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321529.stripped.N4.nii.gz
+sub-CC120640,F,26.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120640.stripped.N4.nii.gz
+ADNI460665,M,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I460665_073_S_4795.stripped.N4.nii.gz
+fcon1000AnnArbor_sub64831,M,19.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub64831.stripped.N4.nii.gz
+ADNI10349528,F,58.8,adni_storage/ADNI_nii_gz_bias_corrected/I10349528_035_S_7004.stripped.N4.nii.gz
+sub-CC221580,F,31.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221580.stripped.N4.nii.gz
+ADNI254253,M,62.3,adni_storage/ADNI_nii_gz_bias_corrected/I254253_116_S_4195.stripped.N4.nii.gz
+sub-CC410323,F,51.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410323.stripped.N4.nii.gz
+DLBS0028464_session_1,F,82.43,dlbs_storage/DLBS_bias_corrected/0028464_session_1.stripped.N4.nii.gz
+fcon1000Milwaukee_sub14692,F,53.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub14692.stripped.N4.nii.gz
+fcon1000Cambridge_sub82213,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub82213.stripped.N4.nii.gz
+fcon1000Bangor_sub46870,M,28.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub46870.stripped.N4.nii.gz
+sub-CC220203,F,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220203.stripped.N4.nii.gz
+fcon1000Cleveland_sub85091,F,52.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub85091.stripped.N4.nii.gz
+sub-CC310051,F,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310051.stripped.N4.nii.gz
+sub-CC120182,M,26.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120182.stripped.N4.nii.gz
+ADNI335130,M,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I335130_073_S_4552.stripped.N4.nii.gz
+sub-CC510483,F,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510483.stripped.N4.nii.gz
+fcon1000NewYork_sub13384,F,28.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub13384.stripped.N4.nii.gz
+DLBS0028327_session_1,F,73.29,dlbs_storage/DLBS_bias_corrected/0028327_session_1.stripped.N4.nii.gz
+ADNI10666738,F,59.9,adni_storage/ADNI_nii_gz_bias_corrected/I10666738_011_S_10045.stripped.N4.nii.gz
+DLBS0028580_session_1,F,60.28,dlbs_storage/DLBS_bias_corrected/0028580_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub86115,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub86115.stripped.N4.nii.gz
+sub-CC410040,M,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410040.stripped.N4.nii.gz
+ADNI368463,M,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I368463_073_S_4552.stripped.N4.nii.gz
+sub-CC520042,M,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520042.stripped.N4.nii.gz
+fcon1000Cambridge_sub53615,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub53615.stripped.N4.nii.gz
+DLBS0028507_session_1,M,36.91,dlbs_storage/DLBS_bias_corrected/0028507_session_1.stripped.N4.nii.gz
+sub-CC510434,F,65.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510434.stripped.N4.nii.gz
+fcon1000Beijing_sub82980,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub82980.stripped.N4.nii.gz
+fcon1000NewYork_sub30247,F,24.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub30247.stripped.N4.nii.gz
+ADNI37606,F,60.1,adni_storage/ADNI_nii_gz_bias_corrected/I37606_051_S_1072.stripped.N4.nii.gz
+DLBS0028562_session_1,F,24.29,dlbs_storage/DLBS_bias_corrected/0028562_session_1.stripped.N4.nii.gz
+DLBS0028543_session_1,F,81.28,dlbs_storage/DLBS_bias_corrected/0028543_session_1.stripped.N4.nii.gz
+fcon1000Cleveland_sub22736,F,28.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub22736.stripped.N4.nii.gz
+sub-CC420198,M,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420198.stripped.N4.nii.gz
+sub-CC410248,M,54.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410248.stripped.N4.nii.gz
+sub-CC121317,F,25.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121317.stripped.N4.nii.gz
+ADNI417674,M,62.0,adni_storage/ADNI_nii_gz_bias_corrected/I417674_128_S_4599.stripped.N4.nii.gz
+sub-CC220806,F,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220806.stripped.N4.nii.gz
+DLBS0028419_session_1,F,78.87,dlbs_storage/DLBS_bias_corrected/0028419_session_1.stripped.N4.nii.gz
+DLBS0028356_session_1,F,43.21,dlbs_storage/DLBS_bias_corrected/0028356_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub40037,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub40037.stripped.N4.nii.gz
+fcon1000Beijing_sub32517,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub32517.stripped.N4.nii.gz
+ADNI1118422,F,59.3,adni_storage/ADNI_nii_gz_bias_corrected/I1118422_006_S_6651.stripped.N4.nii.gz
+sub-CC221935,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221935.stripped.N4.nii.gz
+ADNI958916,F,55.6,adni_storage/ADNI_nii_gz_bias_corrected/I958916_070_S_6191.stripped.N4.nii.gz
+sub-CC420182,F,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420182.stripped.N4.nii.gz
+ADNI335038,M,63.4,adni_storage/ADNI_nii_gz_bias_corrected/I335038_116_S_4195.stripped.N4.nii.gz
+sub-CC520480,M,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520480.stripped.N4.nii.gz
+fcon1000Berlin_sub85681,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub85681.stripped.N4.nii.gz
+fcon1000Cambridge_sub07798,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub07798.stripped.N4.nii.gz
+fcon1000NewYork_sub02503,M,9.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub02503.stripped.N4.nii.gz
+ADNI1076708,F,55.2,adni_storage/ADNI_nii_gz_bias_corrected/I1076708_037_S_6620.stripped.N4.nii.gz
+sub-CC121194,F,24.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121194.stripped.N4.nii.gz
+sub-CC420148,F,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420148.stripped.N4.nii.gz
+sub-CC220535,M,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220535.stripped.N4.nii.gz
+sub-CC110033,M,24.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110033.stripped.N4.nii.gz
+fcon1000Beijing_sub17586,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub17586.stripped.N4.nii.gz
+ADNI1043769,F,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I1043769_003_S_6490.stripped.N4.nii.gz
+sub-CC223115,M,36.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC223115.stripped.N4.nii.gz
+sub-CC510039,F,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510039.stripped.N4.nii.gz
+DLBS0028598_session_1,F,84.06,dlbs_storage/DLBS_bias_corrected/0028598_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub29044,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub29044.stripped.N4.nii.gz
+DLBS0028606_session_1,F,81.91,dlbs_storage/DLBS_bias_corrected/0028606_session_1.stripped.N4.nii.gz
+fcon1000NewYork_sub28808,F,23.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub28808.stripped.N4.nii.gz
+DLBS0028624_session_1,F,81.96,dlbs_storage/DLBS_bias_corrected/0028624_session_1.stripped.N4.nii.gz
+sub-CC510321,F,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510321.stripped.N4.nii.gz
+ADNI358814,F,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I358814_019_S_5019.stripped.N4.nii.gz
+ADNI248875,F,56.3,adni_storage/ADNI_nii_gz_bias_corrected/I248875_031_S_4024.stripped.N4.nii.gz
+sub-CC420259,F,51.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420259.stripped.N4.nii.gz
+ADNI1007384,F,56.2,adni_storage/ADNI_nii_gz_bias_corrected/I1007384_035_S_6380.stripped.N4.nii.gz
+DLBS0028536_session_1,M,82.57,dlbs_storage/DLBS_bias_corrected/0028536_session_1.stripped.N4.nii.gz
+DLBS0028365_session_1,F,22.82,dlbs_storage/DLBS_bias_corrected/0028365_session_1.stripped.N4.nii.gz
+sub-CC610052,M,77.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610052.stripped.N4.nii.gz
+fcon1000Beijing_sub14238,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub14238.stripped.N4.nii.gz
+fcon1000Beijing_sub48676,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub48676.stripped.N4.nii.gz
+sub-CC220610,F,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220610.stripped.N4.nii.gz
+sub-CC610308,M,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610308.stripped.N4.nii.gz
+ADNI10236670,F,61.5,adni_storage/ADNI_nii_gz_bias_corrected/I10236670_168_S_6085.stripped.N4.nii.gz
+fcon1000Milwaukee_sub39259,F,53.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub39259.stripped.N4.nii.gz
+sub-CC520585,F,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520585.stripped.N4.nii.gz
+DLBS0028512_session_1,F,51.64,dlbs_storage/DLBS_bias_corrected/0028512_session_1.stripped.N4.nii.gz
+ADNI944156,F,62.2,adni_storage/ADNI_nii_gz_bias_corrected/I944156_116_S_6129.stripped.N4.nii.gz
+ADNI168100,F,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I168100_051_S_1072.stripped.N4.nii.gz
+ADNI442430,F,59.8,adni_storage/ADNI_nii_gz_bias_corrected/I442430_041_S_4874.stripped.N4.nii.gz
+fcon1000Cambridge_sub53059,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub53059.stripped.N4.nii.gz
+sub-CC420383,F,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420383.stripped.N4.nii.gz
+sub-CC321504,M,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321504.stripped.N4.nii.gz
+fcon1000Beijing_sub41170,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub41170.stripped.N4.nii.gz
+DLBS0028640_session_1,F,33.41,dlbs_storage/DLBS_bias_corrected/0028640_session_1.stripped.N4.nii.gz
+ADNI1399348,F,63.4,adni_storage/ADNI_nii_gz_bias_corrected/I1399348_305_S_6845.stripped.N4.nii.gz
+fcon1000Beijing_sub56757,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub56757.stripped.N4.nii.gz
+ADNI105505,F,63.9,adni_storage/ADNI_nii_gz_bias_corrected/I105505_067_S_1253.stripped.N4.nii.gz
+fcon1000Cambridge_sub07413,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub07413.stripped.N4.nii.gz
+sub-CC610028,M,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610028.stripped.N4.nii.gz
+DLBS0028563_session_1,F,44.97,dlbs_storage/DLBS_bias_corrected/0028563_session_1.stripped.N4.nii.gz
+sub-CC321585,F,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321585.stripped.N4.nii.gz
+fcon1000Atlanta_sub59806,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub59806.stripped.N4.nii.gz
+sub-CC221527,F,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221527.stripped.N4.nii.gz
+ADNI1573570,F,63.1,adni_storage/ADNI_nii_gz_bias_corrected/I1573570_131_S_6805.stripped.N4.nii.gz
+fcon1000AnnArbor_sub56686,F,20.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub56686.stripped.N4.nii.gz
+ADNI10936993,F,55.7,adni_storage/ADNI_nii_gz_bias_corrected/I10936993_033_S_10215.stripped.N4.nii.gz
+fcon1000Cleveland_sub19005,M,29.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub19005.stripped.N4.nii.gz
+fcon1000Atlanta_sub76280,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub76280.stripped.N4.nii.gz
+sub-CC120409,M,18.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120409.stripped.N4.nii.gz
+fcon1000Beijing_sub22595,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub22595.stripped.N4.nii.gz
+fcon1000Leiden_sub13537,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub13537.stripped.N4.nii.gz
+ADNI1344108,F,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I1344108_123_S_6888.stripped.N4.nii.gz
+sub-CC210148,F,30.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210148.stripped.N4.nii.gz
+fcon1000NewYork_sub20732,M,11.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub20732.stripped.N4.nii.gz
+fcon1000Cambridge_sub04187,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub04187.stripped.N4.nii.gz
+ADNI912447,F,63.1,adni_storage/ADNI_nii_gz_bias_corrected/I912447_003_S_6067.stripped.N4.nii.gz
+sub-CC710551,M,85.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710551.stripped.N4.nii.gz
+fcon1000Cambridge_sub93488,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub93488.stripped.N4.nii.gz
+fcon1000Beijing_sub61961,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub61961.stripped.N4.nii.gz
+ADNI280814,F,63.9,adni_storage/ADNI_nii_gz_bias_corrected/I280814_023_S_4448.stripped.N4.nii.gz
+sub-CC321154,M,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321154.stripped.N4.nii.gz
+sub-CC410297,F,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410297.stripped.N4.nii.gz
+sub-CC221565,F,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221565.stripped.N4.nii.gz
+sub-CC510480,M,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510480.stripped.N4.nii.gz
+fcon1000AnnArbor_sub57196,F,77.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub57196.stripped.N4.nii.gz
+ADNI313455,M,60.3,adni_storage/ADNI_nii_gz_bias_corrected/I313455_128_S_4599.stripped.N4.nii.gz
+DLBS0028610_session_1,M,80.82,dlbs_storage/DLBS_bias_corrected/0028610_session_1.stripped.N4.nii.gz
+DLBS0028579_session_1,F,21.77,dlbs_storage/DLBS_bias_corrected/0028579_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub30250,F,19.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub30250.stripped.N4.nii.gz
+fcon1000Beijing_sub52044,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub52044.stripped.N4.nii.gz
+sub-CC620518,M,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620518.stripped.N4.nii.gz
+ADNI1330894,F,62.0,adni_storage/ADNI_nii_gz_bias_corrected/I1330894_305_S_6881.stripped.N4.nii.gz
+fcon1000Leiden_sub58194,M,27.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub58194.stripped.N4.nii.gz
+DLBS0028467_session_1,F,34.4,dlbs_storage/DLBS_bias_corrected/0028467_session_1.stripped.N4.nii.gz
+sub-CC620560,M,72.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620560.stripped.N4.nii.gz
+ADNI1008024,F,59.5,adni_storage/ADNI_nii_gz_bias_corrected/I1008024_177_S_6408.stripped.N4.nii.gz
+ADNI1558656,F,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I1558656_014_S_6076.stripped.N4.nii.gz
+sub-CC710462,F,82.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710462.stripped.N4.nii.gz
+DLBS0028525_session_1,M,60.74,dlbs_storage/DLBS_bias_corrected/0028525_session_1.stripped.N4.nii.gz
+sub-CC721894,F,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721894.stripped.N4.nii.gz
+fcon1000Beijing_sub81074,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub81074.stripped.N4.nii.gz
+sub-CC121397,M,27.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121397.stripped.N4.nii.gz
+ADNI1529914,F,59.4,adni_storage/ADNI_nii_gz_bias_corrected/I1529914_067_S_7033.stripped.N4.nii.gz
+fcon1000Cambridge_sub02591,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub02591.stripped.N4.nii.gz
+fcon1000Baltimore_sub19738,F,27.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub19738.stripped.N4.nii.gz
+fcon1000Cleveland_sub50092,M,56.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub50092.stripped.N4.nii.gz
+sub-CC610101,M,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610101.stripped.N4.nii.gz
+ADNI378273,M,58.1,adni_storage/ADNI_nii_gz_bias_corrected/I378273_014_S_4039.stripped.N4.nii.gz
+ADNI358820,F,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I358820_019_S_5019.stripped.N4.nii.gz
+fcon1000NewYork_sub47087,M,25.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub47087.stripped.N4.nii.gz
+sub-CC620720,F,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620720.stripped.N4.nii.gz
+sub-CC321464,M,45.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321464.stripped.N4.nii.gz
+sub-CC721224,F,87.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721224.stripped.N4.nii.gz
+ADNI1142379,F,61.5,adni_storage/ADNI_nii_gz_bias_corrected/I1142379_022_S_6013.stripped.N4.nii.gz
+ADNI10461527,F,60.0,adni_storage/ADNI_nii_gz_bias_corrected/I10461527_022_S_10040.stripped.N4.nii.gz
+ADNI270531,M,56.5,adni_storage/ADNI_nii_gz_bias_corrected/I270531_014_S_4039.stripped.N4.nii.gz
+fcon1000Berlin_sub18913,F,29.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub18913.stripped.N4.nii.gz
+ADNI65946,M,62.7,adni_storage/ADNI_nii_gz_bias_corrected/I65946_067_S_1185.stripped.N4.nii.gz
+fcon1000Bangor_sub68050,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub68050.stripped.N4.nii.gz
+ADNI1609767,F,61.1,adni_storage/ADNI_nii_gz_bias_corrected/I1609767_073_S_7082.stripped.N4.nii.gz
+fcon1000Cambridge_sub98528,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub98528.stripped.N4.nii.gz
+fcon1000NewYork_sub05208,F,23.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub05208.stripped.N4.nii.gz
+DLBS0028560_session_1,F,57.03,dlbs_storage/DLBS_bias_corrected/0028560_session_1.stripped.N4.nii.gz
+DLBS0028394_session_1,M,20.74,dlbs_storage/DLBS_bias_corrected/0028394_session_1.stripped.N4.nii.gz
+ADNI1327480,F,60.4,adni_storage/ADNI_nii_gz_bias_corrected/I1327480_177_S_6409.stripped.N4.nii.gz
+fcon1000NewYork_sub30860,F,35.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub30860.stripped.N4.nii.gz
+ADNI1516245,F,58.2,adni_storage/ADNI_nii_gz_bias_corrected/I1516245_082_S_6629.stripped.N4.nii.gz
+ADNI393944,F,63.6,adni_storage/ADNI_nii_gz_bias_corrected/I393944_073_S_5023.stripped.N4.nii.gz
+ADNI1493847,M,61.7,adni_storage/ADNI_nii_gz_bias_corrected/I1493847_016_S_7002.stripped.N4.nii.gz
+sub-CC722216,F,86.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC722216.stripped.N4.nii.gz
+fcon1000Cleveland_sub13495,F,45.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub13495.stripped.N4.nii.gz
+fcon1000Beijing_sub52259,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub52259.stripped.N4.nii.gz
+ADNI1183115,M,62.2,adni_storage/ADNI_nii_gz_bias_corrected/I1183115_305_S_6744.stripped.N4.nii.gz
+ADNI393941,F,63.6,adni_storage/ADNI_nii_gz_bias_corrected/I393941_073_S_5023.stripped.N4.nii.gz
+sub-CC110126,F,22.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110126.stripped.N4.nii.gz
+ADNI248872,F,56.3,adni_storage/ADNI_nii_gz_bias_corrected/I248872_031_S_4024.stripped.N4.nii.gz
+DLBS0028373_session_1,F,48.4,dlbs_storage/DLBS_bias_corrected/0028373_session_1.stripped.N4.nii.gz
+ADNI1149157,F,62.8,adni_storage/ADNI_nii_gz_bias_corrected/I1149157_301_S_6698.stripped.N4.nii.gz
+DLBS0028494_session_1,M,74.24,dlbs_storage/DLBS_bias_corrected/0028494_session_1.stripped.N4.nii.gz
+fcon1000Baltimore_sub80221,M,26.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub80221.stripped.N4.nii.gz
+fcon1000AnnArbor_sub18546,M,20.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub18546.stripped.N4.nii.gz
+sub-CC520175,M,61.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520175.stripped.N4.nii.gz
+ADNI950885,M,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I950885_035_S_6160.stripped.N4.nii.gz
+fcon1000Milwaukee_sub56582,F,57.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub56582.stripped.N4.nii.gz
+DLBS0028590_session_1,F,70.79,dlbs_storage/DLBS_bias_corrected/0028590_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub01361,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub01361.stripped.N4.nii.gz
+DLBS0028583_session_1,F,80.54,dlbs_storage/DLBS_bias_corrected/0028583_session_1.stripped.N4.nii.gz
+ADNI344065,F,58.0,adni_storage/ADNI_nii_gz_bias_corrected/I344065_041_S_4874.stripped.N4.nii.gz
+DLBS0028542_session_1,F,69.59,dlbs_storage/DLBS_bias_corrected/0028542_session_1.stripped.N4.nii.gz
+sub-CC320904,M,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320904.stripped.N4.nii.gz
+sub-CC110045,F,24.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110045.stripped.N4.nii.gz
+DLBS0028457_session_1,M,75.67,dlbs_storage/DLBS_bias_corrected/0028457_session_1.stripped.N4.nii.gz
+DLBS0028627_session_1,F,47.3,dlbs_storage/DLBS_bias_corrected/0028627_session_1.stripped.N4.nii.gz
+fcon1000Baltimore_sub54257,F,29.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub54257.stripped.N4.nii.gz
+ADNI884453,F,62.7,adni_storage/ADNI_nii_gz_bias_corrected/I884453_041_S_4874.stripped.N4.nii.gz
+ADNI460668,M,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I460668_073_S_4795.stripped.N4.nii.gz
+fcon1000Beijing_sub17315,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub17315.stripped.N4.nii.gz
+fcon1000Cambridge_sub06037,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub06037.stripped.N4.nii.gz
+fcon1000Cambridge_sub16122,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub16122.stripped.N4.nii.gz
+fcon1000Milwaukee_sub17987,M,47.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub17987.stripped.N4.nii.gz
+sub-CC220132,M,31.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220132.stripped.N4.nii.gz
+ADNI336502,M,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I336502_135_S_4954.stripped.N4.nii.gz
+sub-CC122620,F,24.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC122620.stripped.N4.nii.gz
+sub-CC520398,F,59.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520398.stripped.N4.nii.gz
+ADNI1446304,M,62.2,adni_storage/ADNI_nii_gz_bias_corrected/I1446304_016_S_6941.stripped.N4.nii.gz
+sub-CC210182,F,31.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210182.stripped.N4.nii.gz
+fcon1000Cleveland_sub76139,F,41.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub76139.stripped.N4.nii.gz
+fcon1000Cleveland_sub07835,M,43.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub07835.stripped.N4.nii.gz
+sub-CC310391,M,41.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310391.stripped.N4.nii.gz
+sub-CC510163,F,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510163.stripped.N4.nii.gz
+DLBS0028572_session_1,M,75.37,dlbs_storage/DLBS_bias_corrected/0028572_session_1.stripped.N4.nii.gz
+ADNI359044,M,58.9,adni_storage/ADNI_nii_gz_bias_corrected/I359044_073_S_5090.stripped.N4.nii.gz
+fcon1000Cambridge_sub16390,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub16390.stripped.N4.nii.gz
+sub-CC120319,F,27.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120319.stripped.N4.nii.gz
+DLBS0028491_session_1,F,24.2,dlbs_storage/DLBS_bias_corrected/0028491_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub08816,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub08816.stripped.N4.nii.gz
+ADNI1278681,F,61.7,adni_storage/ADNI_nii_gz_bias_corrected/I1278681_067_S_6529.stripped.N4.nii.gz
+sub-CC221054,M,31.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221054.stripped.N4.nii.gz
+ADNI1017004,M,59.0,adni_storage/ADNI_nii_gz_bias_corrected/I1017004_116_S_6428.stripped.N4.nii.gz
+fcon1000Cambridge_sub84845,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub84845.stripped.N4.nii.gz
+sub-CC620659,F,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620659.stripped.N4.nii.gz
+fcon1000Atlanta_sub07145,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub07145.stripped.N4.nii.gz
+ADNI10435136,M,62.7,adni_storage/ADNI_nii_gz_bias_corrected/I10435136_007_S_10075.stripped.N4.nii.gz
+fcon1000Beijing_sub83728,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub83728.stripped.N4.nii.gz
+fcon1000NewYork_sub29353,F,21.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub29353.stripped.N4.nii.gz
+ADNI388747,M,64.3,adni_storage/ADNI_nii_gz_bias_corrected/I388747_116_S_4195.stripped.N4.nii.gz
+sub-CC221886,M,33.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221886.stripped.N4.nii.gz
+ADNI844180,F,60.4,adni_storage/ADNI_nii_gz_bias_corrected/I844180_022_S_6013.stripped.N4.nii.gz
+ADNI1514928,F,57.9,adni_storage/ADNI_nii_gz_bias_corrected/I1514928_037_S_7006.stripped.N4.nii.gz
+DLBS0028410_session_1,F,74.11,dlbs_storage/DLBS_bias_corrected/0028410_session_1.stripped.N4.nii.gz
+ADNI1516258,F,58.2,adni_storage/ADNI_nii_gz_bias_corrected/I1516258_082_S_6629.stripped.N4.nii.gz
+fcon1000Berlin_sub49134,M,44.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub49134.stripped.N4.nii.gz
+DLBS0028516_session_1,M,88.14,dlbs_storage/DLBS_bias_corrected/0028516_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub92602,M,26.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub92602.stripped.N4.nii.gz
+ADNI274154,M,63.0,adni_storage/ADNI_nii_gz_bias_corrected/I274154_002_S_1155.stripped.N4.nii.gz
+sub-CC520624,F,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520624.stripped.N4.nii.gz
+ADNI1285703,F,59.6,adni_storage/ADNI_nii_gz_bias_corrected/I1285703_131_S_6170.stripped.N4.nii.gz
+fcon1000Atlanta_sub00368,F,28.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub00368.stripped.N4.nii.gz
+fcon1000Cambridge_sub18449,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub18449.stripped.N4.nii.gz
+ADNI1660846,F,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I1660846_019_S_6635.stripped.N4.nii.gz
+sub-CC220901,F,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220901.stripped.N4.nii.gz
+fcon1000Milwaukee_sub76042,F,58.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub76042.stripped.N4.nii.gz
+fcon1000Munchen_sub96591,M,73.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub96591.stripped.N4.nii.gz
+DLBS0028432_session_1,F,53.75,dlbs_storage/DLBS_bias_corrected/0028432_session_1.stripped.N4.nii.gz
+DLBS0028605_session_1,M,63.69,dlbs_storage/DLBS_bias_corrected/0028605_session_1.stripped.N4.nii.gz
+sub-CC520239,F,65.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520239.stripped.N4.nii.gz
+sub-CC120764,M,27.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120764.stripped.N4.nii.gz
+sub-CC620429,M,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620429.stripped.N4.nii.gz
+ADNI132476,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I132476_016_S_1121.stripped.N4.nii.gz
+sub-CC220518,M,30.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220518.stripped.N4.nii.gz
+sub-CC410182,F,53.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410182.stripped.N4.nii.gz
+ADNI365429,F,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I365429_073_S_5023.stripped.N4.nii.gz
+DLBS0028527_session_1,M,23.64,dlbs_storage/DLBS_bias_corrected/0028527_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub05306,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub05306.stripped.N4.nii.gz
+sub-CC710429,M,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710429.stripped.N4.nii.gz
+fcon1000Cambridge_sub20389,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub20389.stripped.N4.nii.gz
+sub-CC120065,F,25.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120065.stripped.N4.nii.gz
+DLBS0028534_session_1,M,54.3,dlbs_storage/DLBS_bias_corrected/0028534_session_1.stripped.N4.nii.gz
+sub-CC710548,M,83.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710548.stripped.N4.nii.gz
+fcon1000Beijing_sub67435,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub67435.stripped.N4.nii.gz
+DLBS0028485_session_1,F,57.23,dlbs_storage/DLBS_bias_corrected/0028485_session_1.stripped.N4.nii.gz
+sub-CC420157,M,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420157.stripped.N4.nii.gz
+sub-CC320698,F,45.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320698.stripped.N4.nii.gz
+ADNI297848,F,57.0,adni_storage/ADNI_nii_gz_bias_corrected/I297848_031_S_4024.stripped.N4.nii.gz
+ADNI415211,F,64.0,adni_storage/ADNI_nii_gz_bias_corrected/I415211_019_S_5019.stripped.N4.nii.gz
+fcon1000Beijing_sub91145,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub91145.stripped.N4.nii.gz
+fcon1000Baltimore_sub23927,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub23927.stripped.N4.nii.gz
+sub-CC110087,F,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110087.stripped.N4.nii.gz
+fcon1000Milwaukee_sub77073,F,52.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub77073.stripped.N4.nii.gz
+sub-CC710099,F,85.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710099.stripped.N4.nii.gz
+DLBS0028619_session_1,M,36.41,dlbs_storage/DLBS_bias_corrected/0028619_session_1.stripped.N4.nii.gz
+ADNI20681,M,64.5,adni_storage/ADNI_nii_gz_bias_corrected/I20681_131_S_0691.stripped.N4.nii.gz
+fcon1000Beijing_sub97442,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub97442.stripped.N4.nii.gz
+DLBS0028528_session_1,M,53.43,dlbs_storage/DLBS_bias_corrected/0028528_session_1.stripped.N4.nii.gz
+ADNI1526394,F,61.4,adni_storage/ADNI_nii_gz_bias_corrected/I1526394_131_S_6170.stripped.N4.nii.gz
+DLBS0028426_session_1,F,82.07,dlbs_storage/DLBS_bias_corrected/0028426_session_1.stripped.N4.nii.gz
+DLBS0028406_session_1,M,47.65,dlbs_storage/DLBS_bias_corrected/0028406_session_1.stripped.N4.nii.gz
+ADNI393947,F,63.6,adni_storage/ADNI_nii_gz_bias_corrected/I393947_073_S_5023.stripped.N4.nii.gz
+sub-CC310385,M,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310385.stripped.N4.nii.gz
+ADNI10932457,M,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I10932457_035_S_7121.stripped.N4.nii.gz
+DLBS0028521_session_1,F,39.74,dlbs_storage/DLBS_bias_corrected/0028521_session_1.stripped.N4.nii.gz
+ADNI1046066,F,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I1046066_041_S_4874.stripped.N4.nii.gz
+fcon1000Milwaukee_sub46312,F,51.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub46312.stripped.N4.nii.gz
+fcon1000Beijing_sub29590,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub29590.stripped.N4.nii.gz
+fcon1000AnnArbor_sub46727,F,13.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub46727.stripped.N4.nii.gz
+sub-CC420888,F,50.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420888.stripped.N4.nii.gz
+sub-CC220234,M,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220234.stripped.N4.nii.gz
+fcon1000Beijing_sub49782,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub49782.stripped.N4.nii.gz
+fcon1000Milwaukee_sub63196,F,49.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub63196.stripped.N4.nii.gz
+sub-CC222956,F,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222956.stripped.N4.nii.gz
+sub-CC610210,M,75.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610210.stripped.N4.nii.gz
+DLBS0028398_session_1,M,83.33,dlbs_storage/DLBS_bias_corrected/0028398_session_1.stripped.N4.nii.gz
+ADNI1151456,M,64.3,adni_storage/ADNI_nii_gz_bias_corrected/I1151456_002_S_6680.stripped.N4.nii.gz
+fcon1000Cambridge_sub90699,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub90699.stripped.N4.nii.gz
+fcon1000Beijing_sub21115,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub21115.stripped.N4.nii.gz
+sub-CC310086,F,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310086.stripped.N4.nii.gz
+sub-CC420173,F,49.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420173.stripped.N4.nii.gz
+DLBS0028571_session_1,M,20.57,dlbs_storage/DLBS_bias_corrected/0028571_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub00306,F,68.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub00306.stripped.N4.nii.gz
+sub-CC222326,M,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222326.stripped.N4.nii.gz
+sub-CC620118,F,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620118.stripped.N4.nii.gz
+ADNI1061954,F,58.0,adni_storage/ADNI_nii_gz_bias_corrected/I1061954_109_S_6364.stripped.N4.nii.gz
+fcon1000Cambridge_sub41773,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub41773.stripped.N4.nii.gz
+sub-CC410390,M,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410390.stripped.N4.nii.gz
+sub-CC610040,M,72.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610040.stripped.N4.nii.gz
+ADNI986502,F,58.4,adni_storage/ADNI_nii_gz_bias_corrected/I986502_094_S_6275.stripped.N4.nii.gz
+fcon1000Leiden_sub04484,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub04484.stripped.N4.nii.gz
+fcon1000Atlanta_sub53122,M,28.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub53122.stripped.N4.nii.gz
+sub-CC321069,F,41.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321069.stripped.N4.nii.gz
+fcon1000Cambridge_sub50454,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub50454.stripped.N4.nii.gz
+ADNI378271,M,58.1,adni_storage/ADNI_nii_gz_bias_corrected/I378271_014_S_4039.stripped.N4.nii.gz
+DLBS0028423_session_1,M,30.69,dlbs_storage/DLBS_bias_corrected/0028423_session_1.stripped.N4.nii.gz
+ADNI10461464,F,56.2,adni_storage/ADNI_nii_gz_bias_corrected/I10461464_033_S_10092.stripped.N4.nii.gz
+fcon1000Beijing_sub51015,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub51015.stripped.N4.nii.gz
+fcon1000Cambridge_sub93269,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub93269.stripped.N4.nii.gz
+fcon1000Beijing_sub50873,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub50873.stripped.N4.nii.gz
+fcon1000Cambridge_sub18295,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub18295.stripped.N4.nii.gz
+ADNI77626,M,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I77626_020_S_1288.stripped.N4.nii.gz
+fcon1000AnnArbor_sub26099,M,73.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub26099.stripped.N4.nii.gz
+DLBS0028430_session_1,F,89.11,dlbs_storage/DLBS_bias_corrected/0028430_session_1.stripped.N4.nii.gz
+sub-CC721291,F,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721291.stripped.N4.nii.gz
+fcon1000Cleveland_sub22935,F,39.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub22935.stripped.N4.nii.gz
+ADNI387580,F,61.8,adni_storage/ADNI_nii_gz_bias_corrected/I387580_036_S_4894.stripped.N4.nii.gz
+DLBS0028592_session_1,F,40.97,dlbs_storage/DLBS_bias_corrected/0028592_session_1.stripped.N4.nii.gz
+ADNI293375,F,56.3,adni_storage/ADNI_nii_gz_bias_corrected/I293375_128_S_4607.stripped.N4.nii.gz
+fcon1000Beijing_sub92799,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub92799.stripped.N4.nii.gz
+DLBS0028354_session_1,M,75.42,dlbs_storage/DLBS_bias_corrected/0028354_session_1.stripped.N4.nii.gz
+sub-CC410287,F,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410287.stripped.N4.nii.gz
+fcon1000NewYork_sub44395,M,21.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub44395.stripped.N4.nii.gz
+ADNI62031,F,56.8,adni_storage/ADNI_nii_gz_bias_corrected/I62031_016_S_1121.stripped.N4.nii.gz
+fcon1000Beijing_sub31058,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub31058.stripped.N4.nii.gz
+ADNI1448643,F,62.9,adni_storage/ADNI_nii_gz_bias_corrected/I1448643_003_S_6954.stripped.N4.nii.gz
+sub-CC320461,M,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320461.stripped.N4.nii.gz
+ADNI1595258,F,53.5,adni_storage/ADNI_nii_gz_bias_corrected/I1595258_037_S_7035.stripped.N4.nii.gz
+fcon1000Beijing_sub29785,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub29785.stripped.N4.nii.gz
+fcon1000Cambridge_sub76631,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub76631.stripped.N4.nii.gz
+ADNI115775,F,59.2,adni_storage/ADNI_nii_gz_bias_corrected/I115775_033_S_0733.stripped.N4.nii.gz
+ADNI1143236,F,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I1143236_023_S_6702.stripped.N4.nii.gz
+DLBS0028459_session_1,M,66.13,dlbs_storage/DLBS_bias_corrected/0028459_session_1.stripped.N4.nii.gz
+ADNI415594,M,57.2,adni_storage/ADNI_nii_gz_bias_corrected/I415594_009_S_5252.stripped.N4.nii.gz
+sub-CC320379,M,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320379.stripped.N4.nii.gz
+fcon1000Beijing_sub08992,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub08992.stripped.N4.nii.gz
+fcon1000Beijing_sub50985,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub50985.stripped.N4.nii.gz
+sub-CC610671,F,70.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610671.stripped.N4.nii.gz
+fcon1000Beijing_sub89742,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub89742.stripped.N4.nii.gz
+DLBS0028397_session_1,F,35.33,dlbs_storage/DLBS_bias_corrected/0028397_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub82352,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub82352.stripped.N4.nii.gz
+fcon1000Beijing_sub64923,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub64923.stripped.N4.nii.gz
+sub-CC321291,F,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321291.stripped.N4.nii.gz
+fcon1000Cambridge_sub69315,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub69315.stripped.N4.nii.gz
+fcon1000Cambridge_sub77598,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub77598.stripped.N4.nii.gz
+sub-CC121685,M,20.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121685.stripped.N4.nii.gz
+sub-CC221244,F,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221244.stripped.N4.nii.gz
+DLBS0028538_session_1,F,76.39,dlbs_storage/DLBS_bias_corrected/0028538_session_1.stripped.N4.nii.gz
+fcon1000Leiden_sub28473,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub28473.stripped.N4.nii.gz
+sub-CC512003,M,62.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC512003.stripped.N4.nii.gz
+sub-CC510609,M,59.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510609.stripped.N4.nii.gz
+ADNI346264,F,61.0,adni_storage/ADNI_nii_gz_bias_corrected/I346264_036_S_4894.stripped.N4.nii.gz
+sub-CC520607,F,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520607.stripped.N4.nii.gz
+fcon1000Beijing_sub51586,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub51586.stripped.N4.nii.gz
+fcon1000Cambridge_sub78552,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub78552.stripped.N4.nii.gz
+fcon1000NewYork_sub59796,M,34.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub59796.stripped.N4.nii.gz
+fcon1000Baltimore_sub29158,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub29158.stripped.N4.nii.gz
+DLBS0028379_session_1,F,73.36,dlbs_storage/DLBS_bias_corrected/0028379_session_1.stripped.N4.nii.gz
+sub-CC320321,F,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320321.stripped.N4.nii.gz
+ADNI395985,F,59.3,adni_storage/ADNI_nii_gz_bias_corrected/I395985_130_S_4982.stripped.N4.nii.gz
+ADNI44672,M,60.1,adni_storage/ADNI_nii_gz_bias_corrected/I44672_020_S_1288.stripped.N4.nii.gz
+fcon1000Cambridge_sub19717,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub19717.stripped.N4.nii.gz
+sub-CC620164,F,72.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620164.stripped.N4.nii.gz
+DLBS0028412_session_1,M,20.89,dlbs_storage/DLBS_bias_corrected/0028412_session_1.stripped.N4.nii.gz
+ADNI285080,M,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I285080_073_S_4552.stripped.N4.nii.gz
+fcon1000Cleveland_sub18011,M,55.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub18011.stripped.N4.nii.gz
+sub-CC620152,M,73.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620152.stripped.N4.nii.gz
+ADNI1525978,F,64.1,adni_storage/ADNI_nii_gz_bias_corrected/I1525978_131_S_6805.stripped.N4.nii.gz
+fcon1000Beijing_sub82426,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub82426.stripped.N4.nii.gz
+DLBS0028438_session_1,M,85.99,dlbs_storage/DLBS_bias_corrected/0028438_session_1.stripped.N4.nii.gz
+ADNI460671,M,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I460671_073_S_4795.stripped.N4.nii.gz
+fcon1000Cambridge_sub20543,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub20543.stripped.N4.nii.gz
+DLBS0028596_session_1,M,36.92,dlbs_storage/DLBS_bias_corrected/0028596_session_1.stripped.N4.nii.gz
+sub-CC420464,F,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420464.stripped.N4.nii.gz
+sub-CC410086,M,57.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410086.stripped.N4.nii.gz
+sub-CC610568,M,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610568.stripped.N4.nii.gz
+DLBS0028622_session_1,F,24.6,dlbs_storage/DLBS_bias_corrected/0028622_session_1.stripped.N4.nii.gz
+ADNI1058589,M,63.4,adni_storage/ADNI_nii_gz_bias_corrected/I1058589_032_S_6602.stripped.N4.nii.gz
+sub-CC420204,F,53.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420204.stripped.N4.nii.gz
+DLBS0028537_session_1,F,62.2,dlbs_storage/DLBS_bias_corrected/0028537_session_1.stripped.N4.nii.gz
+ADNI871844,F,57.7,adni_storage/ADNI_nii_gz_bias_corrected/I871844_141_S_6041.stripped.N4.nii.gz
+ADNI324990,F,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I324990_023_S_4448.stripped.N4.nii.gz
+sub-CC220232,F,36.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220232.stripped.N4.nii.gz
+fcon1000Beijing_sub20127,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub20127.stripped.N4.nii.gz
+fcon1000Bangor_sub48632,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub48632.stripped.N4.nii.gz
+sub-CC720119,M,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720119.stripped.N4.nii.gz
+sub-CC420167,F,51.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420167.stripped.N4.nii.gz
+sub-CC620073,M,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620073.stripped.N4.nii.gz
+fcon1000Cambridge_sub92440,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub92440.stripped.N4.nii.gz
+ADNI1478765,F,56.9,adni_storage/ADNI_nii_gz_bias_corrected/I1478765_037_S_6977.stripped.N4.nii.gz
+sub-CC620106,F,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620106.stripped.N4.nii.gz
+sub-CC710982,M,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710982.stripped.N4.nii.gz
+ADNI1032325,F,56.5,adni_storage/ADNI_nii_gz_bias_corrected/I1032325_070_S_6548.stripped.N4.nii.gz
+DLBS0028453_session_1,F,73.26,dlbs_storage/DLBS_bias_corrected/0028453_session_1.stripped.N4.nii.gz
+DLBS0028473_session_1,F,56.11,dlbs_storage/DLBS_bias_corrected/0028473_session_1.stripped.N4.nii.gz
+fcon1000Newark_sub62985,F,25.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub62985.stripped.N4.nii.gz
+fcon1000AnnArbor_sub45569,F,63.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub45569.stripped.N4.nii.gz
+sub-CC510433,F,61.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510433.stripped.N4.nii.gz
+DLBS0028561_session_1,F,49.51,dlbs_storage/DLBS_bias_corrected/0028561_session_1.stripped.N4.nii.gz
+ADNI1331166,F,57.9,adni_storage/ADNI_nii_gz_bias_corrected/I1331166_116_S_6550.stripped.N4.nii.gz
+sub-CC610392,F,75.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610392.stripped.N4.nii.gz
+DLBS0028359_session_1,F,71.2,dlbs_storage/DLBS_bias_corrected/0028359_session_1.stripped.N4.nii.gz
+DLBS0028404_session_1,F,38.45,dlbs_storage/DLBS_bias_corrected/0028404_session_1.stripped.N4.nii.gz
+ADNI1072842,F,59.9,adni_storage/ADNI_nii_gz_bias_corrected/I1072842_014_S_6076.stripped.N4.nii.gz
+sub-CC310224,F,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310224.stripped.N4.nii.gz
+sub-CC220519,F,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220519.stripped.N4.nii.gz
+sub-CC310160,M,41.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310160.stripped.N4.nii.gz
+sub-CC620567,M,74.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620567.stripped.N4.nii.gz
+DLBS0028638_session_1,F,70.98,dlbs_storage/DLBS_bias_corrected/0028638_session_1.stripped.N4.nii.gz
+fcon1000NewYork_sub31554,F,24.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub31554.stripped.N4.nii.gz
+sub-CC520122,M,67.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520122.stripped.N4.nii.gz
+sub-CC320022,M,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320022.stripped.N4.nii.gz
+DLBS0028391_session_1,M,36.03,dlbs_storage/DLBS_bias_corrected/0028391_session_1.stripped.N4.nii.gz
+DLBS0028506_session_1,F,86.38,dlbs_storage/DLBS_bias_corrected/0028506_session_1.stripped.N4.nii.gz
+fcon1000Munchen_sub26223,M,65.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub26223.stripped.N4.nii.gz
+fcon1000Atlanta_sub15817,M,31.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub15817.stripped.N4.nii.gz
+DLBS0028465_session_1,F,21.67,dlbs_storage/DLBS_bias_corrected/0028465_session_1.stripped.N4.nii.gz
+fcon1000NewYork_sub56734,F,50.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub56734.stripped.N4.nii.gz
+ADNI1191310,M,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I1191310_130_S_6329.stripped.N4.nii.gz
+DLBS0028554_session_1,F,48.67,dlbs_storage/DLBS_bias_corrected/0028554_session_1.stripped.N4.nii.gz
+sub-CC510415,F,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510415.stripped.N4.nii.gz
+fcon1000Newark_sub13411,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub13411.stripped.N4.nii.gz
+fcon1000Cambridge_sub07902,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub07902.stripped.N4.nii.gz
+sub-CC220526,F,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220526.stripped.N4.nii.gz
+sub-CC710037,F,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710037.stripped.N4.nii.gz
+ADNI1072841,F,59.9,adni_storage/ADNI_nii_gz_bias_corrected/I1072841_014_S_6076.stripped.N4.nii.gz
+fcon1000Cambridge_sub60578,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub60578.stripped.N4.nii.gz
+fcon1000Cambridge_sub08723,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub08723.stripped.N4.nii.gz
+DLBS0028535_session_1,F,80.11,dlbs_storage/DLBS_bias_corrected/0028535_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub06272,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub06272.stripped.N4.nii.gz
+ADNI1575205,F,62.6,adni_storage/ADNI_nii_gz_bias_corrected/I1575205_006_S_6651.stripped.N4.nii.gz
+ADNI10666923,F,64.6,adni_storage/ADNI_nii_gz_bias_corrected/I10666923_100_S_10091.stripped.N4.nii.gz
+ADNI323094,F,60.8,adni_storage/ADNI_nii_gz_bias_corrected/I323094_036_S_4894.stripped.N4.nii.gz
+fcon1000Cambridge_sub13902,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub13902.stripped.N4.nii.gz
+fcon1000Cambridge_sub49259,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub49259.stripped.N4.nii.gz
+fcon1000Cambridge_sub65373,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub65373.stripped.N4.nii.gz
+fcon1000Beijing_sub38602,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub38602.stripped.N4.nii.gz
+ADNI400951,M,62.8,adni_storage/ADNI_nii_gz_bias_corrected/I400951_073_S_4795.stripped.N4.nii.gz
+fcon1000Berlin_sub27711,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub27711.stripped.N4.nii.gz
+fcon1000NewYork_sub33581,M,49.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub33581.stripped.N4.nii.gz
+sub-CC120727,F,23.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120727.stripped.N4.nii.gz
+sub-CC620592,F,74.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620592.stripped.N4.nii.gz
+fcon1000Bangor_sub01903,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub01903.stripped.N4.nii.gz
+fcon1000Cambridge_sub53193,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub53193.stripped.N4.nii.gz
+sub-CC220335,M,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220335.stripped.N4.nii.gz
+sub-CC610405,F,72.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610405.stripped.N4.nii.gz
+sub-CC221595,M,33.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221595.stripped.N4.nii.gz
+DLBS0028488_session_1,F,41.33,dlbs_storage/DLBS_bias_corrected/0028488_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub44573,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub44573.stripped.N4.nii.gz
+fcon1000Munchen_sub31272,M,74.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub31272.stripped.N4.nii.gz
+DLBS0028335_session_1,M,71.34,dlbs_storage/DLBS_bias_corrected/0028335_session_1.stripped.N4.nii.gz
+fcon1000Atlanta_sub00354,F,28.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub00354.stripped.N4.nii.gz
+DLBS0028617_session_1,F,50.6,dlbs_storage/DLBS_bias_corrected/0028617_session_1.stripped.N4.nii.gz
+sub-CC520390,F,65.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520390.stripped.N4.nii.gz
+sub-CC620885,M,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620885.stripped.N4.nii.gz
+fcon1000Cambridge_sub31522,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub31522.stripped.N4.nii.gz
+ADNI1185293,F,60.0,adni_storage/ADNI_nii_gz_bias_corrected/I1185293_024_S_6033.stripped.N4.nii.gz
+sub-CC210657,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210657.stripped.N4.nii.gz
+sub-CC321025,M,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321025.stripped.N4.nii.gz
+ADNI381380,M,56.6,adni_storage/ADNI_nii_gz_bias_corrected/I381380_009_S_5252.stripped.N4.nii.gz
+fcon1000Beijing_sub80569,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub80569.stripped.N4.nii.gz
+sub-CC620619,M,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620619.stripped.N4.nii.gz
+sub-CC222258,F,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222258.stripped.N4.nii.gz
+DLBS0028613_session_1,M,74.4,dlbs_storage/DLBS_bias_corrected/0028613_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub66506,F,78.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub66506.stripped.N4.nii.gz
+DLBS0028549_session_1,F,21.45,dlbs_storage/DLBS_bias_corrected/0028549_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub95187,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub95187.stripped.N4.nii.gz
+ADNI1485046,F,58.9,adni_storage/ADNI_nii_gz_bias_corrected/I1485046_116_S_6550.stripped.N4.nii.gz
+sub-CC320574,M,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320574.stripped.N4.nii.gz
+sub-CC620444,F,77.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620444.stripped.N4.nii.gz
+sub-CC510243,M,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510243.stripped.N4.nii.gz
+sub-CC610508,F,77.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610508.stripped.N4.nii.gz
+fcon1000Milwaukee_sub18955,F,53.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub18955.stripped.N4.nii.gz
+sub-CC321557,F,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321557.stripped.N4.nii.gz
+sub-CC510629,M,59.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510629.stripped.N4.nii.gz
+DLBS0028604_session_1,F,22.6,dlbs_storage/DLBS_bias_corrected/0028604_session_1.stripped.N4.nii.gz
+ADNI1227908,F,61.1,adni_storage/ADNI_nii_gz_bias_corrected/I1227908_041_S_6801.stripped.N4.nii.gz
+sub-CC721729,M,81.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721729.stripped.N4.nii.gz
+sub-CC320325,F,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320325.stripped.N4.nii.gz
+sub-CC221511,F,36.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221511.stripped.N4.nii.gz
+sub-CC610076,F,77.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610076.stripped.N4.nii.gz
+sub-CC520980,M,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520980.stripped.N4.nii.gz
+DLBS0028478_session_1,F,70.35,dlbs_storage/DLBS_bias_corrected/0028478_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub46058,M,25.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub46058.stripped.N4.nii.gz
+ADNI228875,F,56.0,adni_storage/ADNI_nii_gz_bias_corrected/I228875_031_S_4024.stripped.N4.nii.gz
+DLBS0028601_session_1,M,71.54,dlbs_storage/DLBS_bias_corrected/0028601_session_1.stripped.N4.nii.gz
+fcon1000Cleveland_sub26557,M,51.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub26557.stripped.N4.nii.gz
+sub-CC410032,M,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410032.stripped.N4.nii.gz
+ADNI1461224,M,61.6,adni_storage/ADNI_nii_gz_bias_corrected/I1461224_070_S_6966.stripped.N4.nii.gz
+sub-CC320336,F,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320336.stripped.N4.nii.gz
+sub-CC620354,F,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620354.stripped.N4.nii.gz
+fcon1000Leiden_sub92061,M,28.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub92061.stripped.N4.nii.gz
+DLBS0028344_session_1,F,69.81,dlbs_storage/DLBS_bias_corrected/0028344_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub95644,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub95644.stripped.N4.nii.gz
+ADNI1215082,F,56.9,adni_storage/ADNI_nii_gz_bias_corrected/I1215082_116_S_6550.stripped.N4.nii.gz
+sub-CC110187,F,25.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110187.stripped.N4.nii.gz
+fcon1000Baltimore_sub96234,F,30.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub96234.stripped.N4.nii.gz
+fcon1000Leiden_sub56299,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub56299.stripped.N4.nii.gz
+ADNI400958,M,62.8,adni_storage/ADNI_nii_gz_bias_corrected/I400958_073_S_4795.stripped.N4.nii.gz
+fcon1000Beijing_sub08001,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub08001.stripped.N4.nii.gz
+fcon1000Leiden_sub93194,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub93194.stripped.N4.nii.gz
+DLBS0028508_session_1,F,28.67,dlbs_storage/DLBS_bias_corrected/0028508_session_1.stripped.N4.nii.gz
+fcon1000Leiden_sub01787,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub01787.stripped.N4.nii.gz
+sub-CC320776,M,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320776.stripped.N4.nii.gz
+sub-CC121111,M,18.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121111.stripped.N4.nii.gz
+fcon1000Cambridge_sub69397,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub69397.stripped.N4.nii.gz
+fcon1000AnnArbor_sub99692,F,66.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub99692.stripped.N4.nii.gz
+fcon1000Newark_sub59799,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub59799.stripped.N4.nii.gz
+sub-CC321368,M,45.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321368.stripped.N4.nii.gz
+ADNI381382,M,56.6,adni_storage/ADNI_nii_gz_bias_corrected/I381382_009_S_5252.stripped.N4.nii.gz
+ADNI90757,F,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I90757_051_S_1072.stripped.N4.nii.gz
+fcon1000AnnArbor_sub57025,F,68.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub57025.stripped.N4.nii.gz
+sub-CC320575,F,39.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320575.stripped.N4.nii.gz
+ADNI323095,F,60.8,adni_storage/ADNI_nii_gz_bias_corrected/I323095_036_S_4894.stripped.N4.nii.gz
+sub-CC710088,F,83.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710088.stripped.N4.nii.gz
+sub-CC710518,F,83.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710518.stripped.N4.nii.gz
+DLBS0028587_session_1,F,72.59,dlbs_storage/DLBS_bias_corrected/0028587_session_1.stripped.N4.nii.gz
+DLBS0028331_session_1,F,77.51,dlbs_storage/DLBS_bias_corrected/0028331_session_1.stripped.N4.nii.gz
+fcon1000Newark_sub58526,F,25.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub58526.stripped.N4.nii.gz
+ADNI362186,F,58.4,adni_storage/ADNI_nii_gz_bias_corrected/I362186_041_S_4874.stripped.N4.nii.gz
+sub-CC510258,M,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510258.stripped.N4.nii.gz
+sub-CC520673,M,67.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520673.stripped.N4.nii.gz
+DLBS0028484_session_1,M,70.08,dlbs_storage/DLBS_bias_corrected/0028484_session_1.stripped.N4.nii.gz
+fcon1000Munchen_sub26670,M,64.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub26670.stripped.N4.nii.gz
+fcon1000Cambridge_sub25044,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub25044.stripped.N4.nii.gz
+fcon1000NewYork_sub17078,M,34.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub17078.stripped.N4.nii.gz
+sub-CC410015,F,51.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410015.stripped.N4.nii.gz
+sub-CC410084,F,57.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410084.stripped.N4.nii.gz
+sub-CC520136,F,65.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520136.stripped.N4.nii.gz
+fcon1000Berlin_sub77281,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub77281.stripped.N4.nii.gz
+fcon1000Milwaukee_sub53971,F,47.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub53971.stripped.N4.nii.gz
+sub-CC621128,M,75.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC621128.stripped.N4.nii.gz
+sub-CC121428,F,26.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121428.stripped.N4.nii.gz
+ADNI10666919,F,59.7,adni_storage/ADNI_nii_gz_bias_corrected/I10666919_301_S_10139.stripped.N4.nii.gz
+sub-CC621199,M,72.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC621199.stripped.N4.nii.gz
+fcon1000Beijing_sub93689,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub93689.stripped.N4.nii.gz
+ADNI1080321,F,62.9,adni_storage/ADNI_nii_gz_bias_corrected/I1080321_114_S_6597.stripped.N4.nii.gz
+DLBS0028434_session_1,F,70.1,dlbs_storage/DLBS_bias_corrected/0028434_session_1.stripped.N4.nii.gz
+DLBS0028576_session_1,F,32.92,dlbs_storage/DLBS_bias_corrected/0028576_session_1.stripped.N4.nii.gz
+DLBS0028342_session_1,F,31.98,dlbs_storage/DLBS_bias_corrected/0028342_session_1.stripped.N4.nii.gz
+ADNI460666,M,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I460666_073_S_4795.stripped.N4.nii.gz
+sub-CC420222,M,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420222.stripped.N4.nii.gz
+sub-CC720685,M,81.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720685.stripped.N4.nii.gz
+sub-CC221209,F,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221209.stripped.N4.nii.gz
+fcon1000Milwaukee_sub09931,M,48.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub09931.stripped.N4.nii.gz
+fcon1000Cambridge_sub14278,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub14278.stripped.N4.nii.gz
+fcon1000Cambridge_sub29425,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub29425.stripped.N4.nii.gz
+ADNI1195772,F,63.7,adni_storage/ADNI_nii_gz_bias_corrected/I1195772_019_S_6483.stripped.N4.nii.gz
+fcon1000Cambridge_sub55874,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub55874.stripped.N4.nii.gz
+ADNI274149,M,63.0,adni_storage/ADNI_nii_gz_bias_corrected/I274149_002_S_1155.stripped.N4.nii.gz
+ADNI10302489,M,58.5,adni_storage/ADNI_nii_gz_bias_corrected/I10302489_011_S_7112.stripped.N4.nii.gz
+fcon1000NewYork_sub23844,F,20.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub23844.stripped.N4.nii.gz
+DLBS0028515_session_1,F,50.9,dlbs_storage/DLBS_bias_corrected/0028515_session_1.stripped.N4.nii.gz
+ADNI1253903,M,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I1253903_032_S_6602.stripped.N4.nii.gz
+fcon1000Beijing_sub06880,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub06880.stripped.N4.nii.gz
+sub-CC120313,M,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120313.stripped.N4.nii.gz
+ADNI362185,F,58.4,adni_storage/ADNI_nii_gz_bias_corrected/I362185_041_S_4874.stripped.N4.nii.gz
+ADNI10874412,M,63.0,adni_storage/ADNI_nii_gz_bias_corrected/I10874412_019_S_10145.stripped.N4.nii.gz
+ADNI395983,F,59.3,adni_storage/ADNI_nii_gz_bias_corrected/I395983_130_S_4982.stripped.N4.nii.gz
+fcon1000Baltimore_sub91622,M,33.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub91622.stripped.N4.nii.gz
+DLBS0028414_session_1,F,68.45,dlbs_storage/DLBS_bias_corrected/0028414_session_1.stripped.N4.nii.gz
+ADNI114056,F,57.9,adni_storage/ADNI_nii_gz_bias_corrected/I114056_016_S_1121.stripped.N4.nii.gz
+sub-CC510395,M,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510395.stripped.N4.nii.gz
+fcon1000Cambridge_sub89107,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub89107.stripped.N4.nii.gz
+fcon1000Cambridge_sub60797,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub60797.stripped.N4.nii.gz
+fcon1000Newark_sub55760,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub55760.stripped.N4.nii.gz
+DLBS0028446_session_1,M,50.89,dlbs_storage/DLBS_bias_corrected/0028446_session_1.stripped.N4.nii.gz
+fcon1000Leiden_sub87320,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub87320.stripped.N4.nii.gz
+DLBS0028608_session_1,F,38.13,dlbs_storage/DLBS_bias_corrected/0028608_session_1.stripped.N4.nii.gz
+sub-CC321331,M,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321331.stripped.N4.nii.gz
+fcon1000Beijing_sub34895,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub34895.stripped.N4.nii.gz
+sub-CC220697,M,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220697.stripped.N4.nii.gz
+ADNI1454373,F,55.8,adni_storage/ADNI_nii_gz_bias_corrected/I1454373_137_S_6906.stripped.N4.nii.gz
+ADNI844181,F,60.4,adni_storage/ADNI_nii_gz_bias_corrected/I844181_022_S_6013.stripped.N4.nii.gz
+sub-CC621184,M,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC621184.stripped.N4.nii.gz
+sub-CC510473,M,65.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510473.stripped.N4.nii.gz
+DLBS0028370_session_1,F,22.88,dlbs_storage/DLBS_bias_corrected/0028370_session_1.stripped.N4.nii.gz
+DLBS0028567_session_1,F,85.6,dlbs_storage/DLBS_bias_corrected/0028567_session_1.stripped.N4.nii.gz
+fcon1000NewYork_sub01912,M,31.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub01912.stripped.N4.nii.gz
+ADNI1469814,M,57.0,adni_storage/ADNI_nii_gz_bias_corrected/I1469814_073_S_6925.stripped.N4.nii.gz
+sub-CC410222,F,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410222.stripped.N4.nii.gz
+DLBS0028445_session_1,M,55.37,dlbs_storage/DLBS_bias_corrected/0028445_session_1.stripped.N4.nii.gz
+DLBS0028451_session_1,F,43.61,dlbs_storage/DLBS_bias_corrected/0028451_session_1.stripped.N4.nii.gz
+DLBS0028503_session_1,F,88.32,dlbs_storage/DLBS_bias_corrected/0028503_session_1.stripped.N4.nii.gz
+sub-CC221220,F,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221220.stripped.N4.nii.gz
+sub-CC310203,F,39.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310203.stripped.N4.nii.gz
+ADNI73490,F,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I73490_067_S_1253.stripped.N4.nii.gz
+fcon1000Beijing_sub04050,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub04050.stripped.N4.nii.gz
+ADNI168101,F,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I168101_051_S_1072.stripped.N4.nii.gz
+sub-CC420094,M,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420094.stripped.N4.nii.gz
+fcon1000Cambridge_sub15172,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub15172.stripped.N4.nii.gz
+fcon1000Cambridge_sub14183,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub14183.stripped.N4.nii.gz
+sub-CC220506,F,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220506.stripped.N4.nii.gz
+sub-CC722542,M,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC722542.stripped.N4.nii.gz
+DLBS0028401_session_1,F,54.05,dlbs_storage/DLBS_bias_corrected/0028401_session_1.stripped.N4.nii.gz
+sub-CC721891,M,83.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721891.stripped.N4.nii.gz
+sub-CC510161,M,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510161.stripped.N4.nii.gz
+sub-CC621011,M,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC621011.stripped.N4.nii.gz
+sub-CC520209,F,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520209.stripped.N4.nii.gz
+DLBS0028625_session_1,F,60.37,dlbs_storage/DLBS_bias_corrected/0028625_session_1.stripped.N4.nii.gz
+ADNI60863,F,57.3,adni_storage/ADNI_nii_gz_bias_corrected/I60863_133_S_0913.stripped.N4.nii.gz
+ADNI1559412,F,60.2,adni_storage/ADNI_nii_gz_bias_corrected/I1559412_082_S_6629.stripped.N4.nii.gz
+sub-CC220920,F,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220920.stripped.N4.nii.gz
+fcon1000Cambridge_sub83409,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub83409.stripped.N4.nii.gz
+ADNI1406523,F,55.5,adni_storage/ADNI_nii_gz_bias_corrected/I1406523_305_S_6899.stripped.N4.nii.gz
+fcon1000Berlin_sub67166,F,32.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub67166.stripped.N4.nii.gz
+sub-CC221336,M,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221336.stripped.N4.nii.gz
+fcon1000Beijing_sub07717,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub07717.stripped.N4.nii.gz
+fcon1000AnnArbor_sub05580,F,77.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub05580.stripped.N4.nii.gz
+DLBS0028386_session_1,F,33.13,dlbs_storage/DLBS_bias_corrected/0028386_session_1.stripped.N4.nii.gz
+ADNI10386604,F,57.2,adni_storage/ADNI_nii_gz_bias_corrected/I10386604_035_S_10067.stripped.N4.nii.gz
+fcon1000Beijing_sub68012,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub68012.stripped.N4.nii.gz
+ADNI1194377,F,55.1,adni_storage/ADNI_nii_gz_bias_corrected/I1194377_016_S_6773.stripped.N4.nii.gz
+sub-CC410113,M,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410113.stripped.N4.nii.gz
+sub-CC120264,M,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120264.stripped.N4.nii.gz
+ADNI347410,F,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I347410_130_S_4997.stripped.N4.nii.gz
+ADNI1254370,F,60.9,adni_storage/ADNI_nii_gz_bias_corrected/I1254370_014_S_6076.stripped.N4.nii.gz
+ADNI423069,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I423069_128_S_4607.stripped.N4.nii.gz
+DLBS0028476_session_1,F,65.91,dlbs_storage/DLBS_bias_corrected/0028476_session_1.stripped.N4.nii.gz
+sub-CC510474,M,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510474.stripped.N4.nii.gz
+fcon1000Milwaukee_sub56333,M,53.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub56333.stripped.N4.nii.gz
+sub-CC221373,M,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221373.stripped.N4.nii.gz
+DLBS0028362_session_1,F,60.87,dlbs_storage/DLBS_bias_corrected/0028362_session_1.stripped.N4.nii.gz
+sub-CC510015,F,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510015.stripped.N4.nii.gz
+sub-CC510438,M,62.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510438.stripped.N4.nii.gz
+ADNI974338,M,62.1,adni_storage/ADNI_nii_gz_bias_corrected/I974338_003_S_6257.stripped.N4.nii.gz
+sub-CC320636,M,45.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320636.stripped.N4.nii.gz
+sub-CC520395,F,61.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520395.stripped.N4.nii.gz
+fcon1000Berlin_sub33248,F,28.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub33248.stripped.N4.nii.gz
+sub-CC210088,F,36.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210088.stripped.N4.nii.gz
+DLBS0028472_session_1,M,42.45,dlbs_storage/DLBS_bias_corrected/0028472_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub39737,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub39737.stripped.N4.nii.gz
+ADNI344059,F,58.0,adni_storage/ADNI_nii_gz_bias_corrected/I344059_041_S_4874.stripped.N4.nii.gz
+fcon1000Milwaukee_sub45852,F,63.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub45852.stripped.N4.nii.gz
+fcon1000Beijing_sub22715,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub22715.stripped.N4.nii.gz
+fcon1000Beijing_sub17093,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub17093.stripped.N4.nii.gz
+fcon1000NewYork_sub28795,M,10.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub28795.stripped.N4.nii.gz
+DLBS0028544_session_1,M,81.75,dlbs_storage/DLBS_bias_corrected/0028544_session_1.stripped.N4.nii.gz
+sub-CC520134,F,67.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520134.stripped.N4.nii.gz
+ADNI10937009,M,59.1,adni_storage/ADNI_nii_gz_bias_corrected/I10937009_052_S_10188.stripped.N4.nii.gz
+ADNI132477,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I132477_016_S_1121.stripped.N4.nii.gz
+sub-CC721519,F,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721519.stripped.N4.nii.gz
+ADNI10454377,M,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I10454377_022_S_10055.stripped.N4.nii.gz
+sub-CC222185,F,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222185.stripped.N4.nii.gz
+ADNI1446679,F,59.8,adni_storage/ADNI_nii_gz_bias_corrected/I1446679_070_S_6942.stripped.N4.nii.gz
+fcon1000AnnArbor_sub85257,M,26.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub85257.stripped.N4.nii.gz
+fcon1000NewYork_sub29216,M,29.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub29216.stripped.N4.nii.gz
+DLBS0028479_session_1,M,40.93,dlbs_storage/DLBS_bias_corrected/0028479_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub35776,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub35776.stripped.N4.nii.gz
+sub-CC210023,M,31.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210023.stripped.N4.nii.gz
+ADNI267809,F,64.1,adni_storage/ADNI_nii_gz_bias_corrected/I267809_037_S_4028.stripped.N4.nii.gz
+fcon1000NewYork_sub53461,M,27.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub53461.stripped.N4.nii.gz
+fcon1000Milwaukee_sub55176,F,63.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub55176.stripped.N4.nii.gz
+DLBS0028420_session_1,F,81.2,dlbs_storage/DLBS_bias_corrected/0028420_session_1.stripped.N4.nii.gz
+sub-CC510237,M,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510237.stripped.N4.nii.gz
+fcon1000Cambridge_sub49998,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub49998.stripped.N4.nii.gz
+fcon1000Cambridge_sub13187,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub13187.stripped.N4.nii.gz
+sub-CC222496,M,38.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222496.stripped.N4.nii.gz
+fcon1000Cambridge_sub04491,F,25.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub04491.stripped.N4.nii.gz
+sub-CC120376,F,18.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120376.stripped.N4.nii.gz
+ADNI114263,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I114263_133_S_0913.stripped.N4.nii.gz
+DLBS0028602_session_1,F,29.19,dlbs_storage/DLBS_bias_corrected/0028602_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub15846,F,70.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub15846.stripped.N4.nii.gz
+fcon1000Baltimore_sub76160,M,34.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub76160.stripped.N4.nii.gz
+fcon1000Cambridge_sub99462,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub99462.stripped.N4.nii.gz
+fcon1000Beijing_sub66889,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub66889.stripped.N4.nii.gz
+ADNI87682,F,57.8,adni_storage/ADNI_nii_gz_bias_corrected/I87682_133_S_0913.stripped.N4.nii.gz
+ADNI1254369,F,60.9,adni_storage/ADNI_nii_gz_bias_corrected/I1254369_014_S_6076.stripped.N4.nii.gz
+ADNI367180,M,61.8,adni_storage/ADNI_nii_gz_bias_corrected/I367180_135_S_4954.stripped.N4.nii.gz
+fcon1000Berlin_sub54976,M,37.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub54976.stripped.N4.nii.gz
+fcon1000Beijing_sub87776,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub87776.stripped.N4.nii.gz
+fcon1000Milwaukee_sub59359,F,51.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub59359.stripped.N4.nii.gz
+ADNI1079629,F,57.0,adni_storage/ADNI_nii_gz_bias_corrected/I1079629_082_S_6629.stripped.N4.nii.gz
+sub-CC221352,F,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221352.stripped.N4.nii.gz
+sub-CC210124,F,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210124.stripped.N4.nii.gz
+fcon1000NewYork_sub19579,M,46.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub19579.stripped.N4.nii.gz
+sub-CC320568,F,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320568.stripped.N4.nii.gz
+sub-CC520503,M,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520503.stripped.N4.nii.gz
+fcon1000Baltimore_sub77572,F,27.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub77572.stripped.N4.nii.gz
+DLBS0028375_session_1,F,59.07,dlbs_storage/DLBS_bias_corrected/0028375_session_1.stripped.N4.nii.gz
+fcon1000Milwaukee_sub30157,M,54.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub30157.stripped.N4.nii.gz
+fcon1000AnnArbor_sub78151,M,23.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub78151.stripped.N4.nii.gz
+DLBS0028586_session_1,M,79.85,dlbs_storage/DLBS_bias_corrected/0028586_session_1.stripped.N4.nii.gz
+DLBS0028447_session_1,M,51.78,dlbs_storage/DLBS_bias_corrected/0028447_session_1.stripped.N4.nii.gz
+sub-CC221954,F,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221954.stripped.N4.nii.gz
+sub-CC320893,F,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320893.stripped.N4.nii.gz
+fcon1000Beijing_sub07144,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub07144.stripped.N4.nii.gz
+DLBS0028482_session_1,M,63.47,dlbs_storage/DLBS_bias_corrected/0028482_session_1.stripped.N4.nii.gz
+DLBS0028349_session_1,M,64.17,dlbs_storage/DLBS_bias_corrected/0028349_session_1.stripped.N4.nii.gz
+fcon1000Munchen_sub81737,F,67.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub81737.stripped.N4.nii.gz
+ADNI341153,M,60.6,adni_storage/ADNI_nii_gz_bias_corrected/I341153_128_S_4599.stripped.N4.nii.gz
+sub-CC110098,M,23.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110098.stripped.N4.nii.gz
+sub-CC621118,M,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC621118.stripped.N4.nii.gz
+DLBS0028452_session_1,F,59.71,dlbs_storage/DLBS_bias_corrected/0028452_session_1.stripped.N4.nii.gz
+ADNI228845,F,63.5,adni_storage/ADNI_nii_gz_bias_corrected/I228845_037_S_4028.stripped.N4.nii.gz
+DLBS0028614_session_1,F,53.77,dlbs_storage/DLBS_bias_corrected/0028614_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub24670,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub24670.stripped.N4.nii.gz
+ADNI1181048,M,60.0,adni_storage/ADNI_nii_gz_bias_corrected/I1181048_116_S_6428.stripped.N4.nii.gz
+fcon1000Cambridge_sub59729,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub59729.stripped.N4.nii.gz
+DLBS0028425_session_1,M,23.02,dlbs_storage/DLBS_bias_corrected/0028425_session_1.stripped.N4.nii.gz
+sub-CC510342,F,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510342.stripped.N4.nii.gz
+fcon1000Milwaukee_sub58967,F,61.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub58967.stripped.N4.nii.gz
+sub-CC220107,M,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220107.stripped.N4.nii.gz
+sub-CC620499,M,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620499.stripped.N4.nii.gz
+fcon1000Cambridge_sub89435,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub89435.stripped.N4.nii.gz
+sub-CC420162,F,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420162.stripped.N4.nii.gz
+fcon1000Newark_sub78225,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub78225.stripped.N4.nii.gz
+fcon1000AnnArbor_sub38614,M,14.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub38614.stripped.N4.nii.gz
+fcon1000Leiden_sub52853,M,27.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub52853.stripped.N4.nii.gz
+DLBS0028353_session_1,M,60.01,dlbs_storage/DLBS_bias_corrected/0028353_session_1.stripped.N4.nii.gz
+ADNI10471094,F,58.8,adni_storage/ADNI_nii_gz_bias_corrected/I10471094_073_S_10113.stripped.N4.nii.gz
+sub-CC312058,M,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC312058.stripped.N4.nii.gz
+sub-CC710176,F,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710176.stripped.N4.nii.gz
+ADNI296147,M,63.9,adni_storage/ADNI_nii_gz_bias_corrected/I296147_116_S_4625.stripped.N4.nii.gz
+sub-CC520562,M,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520562.stripped.N4.nii.gz
+DLBS0028621_session_1,F,30.6,dlbs_storage/DLBS_bias_corrected/0028621_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub53269,M,21.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub53269.stripped.N4.nii.gz
+DLBS0028529_session_1,F,86.77,dlbs_storage/DLBS_bias_corrected/0028529_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub30272,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub30272.stripped.N4.nii.gz
+fcon1000Beijing_sub89088,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub89088.stripped.N4.nii.gz
+DLBS0028396_session_1,M,67.74,dlbs_storage/DLBS_bias_corrected/0028396_session_1.stripped.N4.nii.gz
+sub-CC620479,M,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620479.stripped.N4.nii.gz
+ADNI352951,M,64.0,adni_storage/ADNI_nii_gz_bias_corrected/I352951_002_S_1155.stripped.N4.nii.gz
+fcon1000Milwaukee_sub64463,M,52.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub64463.stripped.N4.nii.gz
+sub-CC620129,M,75.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620129.stripped.N4.nii.gz
+DLBS0028594_session_1,M,80.29,dlbs_storage/DLBS_bias_corrected/0028594_session_1.stripped.N4.nii.gz
+sub-CC220974,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220974.stripped.N4.nii.gz
+sub-CC321281,F,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321281.stripped.N4.nii.gz
+sub-CC310252,M,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310252.stripped.N4.nii.gz
+ADNI944422,M,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I944422_041_S_6136.stripped.N4.nii.gz
+fcon1000Cambridge_sub61753,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub61753.stripped.N4.nii.gz
+sub-CC310129,M,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310129.stripped.N4.nii.gz
+sub-CC320088,M,41.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320088.stripped.N4.nii.gz
+sub-CC621642,M,73.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC621642.stripped.N4.nii.gz
+fcon1000Cambridge_sub17772,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub17772.stripped.N4.nii.gz
+sub-CC321174,F,39.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321174.stripped.N4.nii.gz
+ADNI1486321,M,57.1,adni_storage/ADNI_nii_gz_bias_corrected/I1486321_941_S_6998.stripped.N4.nii.gz
+ADNI313534,M,64.1,adni_storage/ADNI_nii_gz_bias_corrected/I313534_116_S_4625.stripped.N4.nii.gz
+DLBS0028340_session_1,F,67.83,dlbs_storage/DLBS_bias_corrected/0028340_session_1.stripped.N4.nii.gz
+sub-CC321431,F,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321431.stripped.N4.nii.gz
+sub-CC711158,M,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC711158.stripped.N4.nii.gz
+sub-CC420412,M,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420412.stripped.N4.nii.gz
+sub-CC510115,F,62.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510115.stripped.N4.nii.gz
+sub-CC120208,F,24.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120208.stripped.N4.nii.gz
+fcon1000NewYork_sub41546,F,24.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub41546.stripped.N4.nii.gz
+sub-CC610372,M,70.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610372.stripped.N4.nii.gz
+DLBS0028350_session_1,F,49.82,dlbs_storage/DLBS_bias_corrected/0028350_session_1.stripped.N4.nii.gz
+sub-CC620785,M,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620785.stripped.N4.nii.gz
+sub-CC710382,M,83.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710382.stripped.N4.nii.gz
+DLBS0028352_session_1,F,29.84,dlbs_storage/DLBS_bias_corrected/0028352_session_1.stripped.N4.nii.gz
+ADNI296878,F,58.5,adni_storage/ADNI_nii_gz_bias_corrected/I296878_012_S_4642.stripped.N4.nii.gz
+fcon1000Munchen_sub96752,F,63.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub96752.stripped.N4.nii.gz
+ADNI1514331,F,52.8,adni_storage/ADNI_nii_gz_bias_corrected/I1514331_037_S_6992.stripped.N4.nii.gz
+ADNI368465,M,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I368465_073_S_4552.stripped.N4.nii.gz
+fcon1000Baltimore_sub90893,F,39.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub90893.stripped.N4.nii.gz
+fcon1000Cambridge_sub82435,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub82435.stripped.N4.nii.gz
+fcon1000Beijing_sub15441,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub15441.stripped.N4.nii.gz
+sub-CC621284,M,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC621284.stripped.N4.nii.gz
+sub-CC410354,M,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410354.stripped.N4.nii.gz
+sub-CC122405,M,27.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC122405.stripped.N4.nii.gz
+fcon1000Cambridge_sub50272,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub50272.stripped.N4.nii.gz
+sub-CC721707,F,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721707.stripped.N4.nii.gz
+fcon1000Bangor_sub04097,M,36.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub04097.stripped.N4.nii.gz
+ADNI1092240,M,57.4,adni_storage/ADNI_nii_gz_bias_corrected/I1092240_114_S_6039.stripped.N4.nii.gz
+DLBS0028631_session_1,F,29.92,dlbs_storage/DLBS_bias_corrected/0028631_session_1.stripped.N4.nii.gz
+fcon1000Milwaukee_sub87784,M,58.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub87784.stripped.N4.nii.gz
+ADNI365455,M,62.1,adni_storage/ADNI_nii_gz_bias_corrected/I365455_073_S_4795.stripped.N4.nii.gz
+sub-CC310414,M,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310414.stripped.N4.nii.gz
+fcon1000Cleveland_sub02480,F,56.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub02480.stripped.N4.nii.gz
+fcon1000Milwaukee_sub24237,F,52.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub24237.stripped.N4.nii.gz
+fcon1000NewYork_sub15213,M,16.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub15213.stripped.N4.nii.gz
+sub-CC110037,M,18.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110037.stripped.N4.nii.gz
+fcon1000Beijing_sub71693,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub71693.stripped.N4.nii.gz
+DLBS0028490_session_1,F,63.73,dlbs_storage/DLBS_bias_corrected/0028490_session_1.stripped.N4.nii.gz
+sub-CC321880,F,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321880.stripped.N4.nii.gz
+sub-CC321594,M,43.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321594.stripped.N4.nii.gz
+sub-CC221828,F,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221828.stripped.N4.nii.gz
+DLBS0028474_session_1,F,83.7,dlbs_storage/DLBS_bias_corrected/0028474_session_1.stripped.N4.nii.gz
+ADNI247842,F,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I247842_037_S_4028.stripped.N4.nii.gz
+fcon1000Cambridge_sub57221,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub57221.stripped.N4.nii.gz
+sub-CC120286,M,22.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120286.stripped.N4.nii.gz
+DLBS0028441_session_1,M,45.18,dlbs_storage/DLBS_bias_corrected/0028441_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub28698,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub28698.stripped.N4.nii.gz
+ADNI117521,F,61.7,adni_storage/ADNI_nii_gz_bias_corrected/I117521_051_S_1072.stripped.N4.nii.gz
+fcon1000Cambridge_sub47162,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub47162.stripped.N4.nii.gz
+fcon1000Beijing_sub54890,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub54890.stripped.N4.nii.gz
+sub-CC720304,F,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720304.stripped.N4.nii.gz
+fcon1000Cambridge_sub58682,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub58682.stripped.N4.nii.gz
+fcon1000Beijing_sub05267,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub05267.stripped.N4.nii.gz
+ADNI36695,F,56.4,adni_storage/ADNI_nii_gz_bias_corrected/I36695_016_S_1121.stripped.N4.nii.gz
+sub-CC520775,F,61.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520775.stripped.N4.nii.gz
+fcon1000Beijing_sub81062,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub81062.stripped.N4.nii.gz
+fcon1000Milwaukee_sub58677,F,46.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub58677.stripped.N4.nii.gz
+DLBS0028374_session_1,F,55.62,dlbs_storage/DLBS_bias_corrected/0028374_session_1.stripped.N4.nii.gz
+DLBS0028345_session_1,F,59.73,dlbs_storage/DLBS_bias_corrected/0028345_session_1.stripped.N4.nii.gz
+DLBS0028381_session_1,F,24.24,dlbs_storage/DLBS_bias_corrected/0028381_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub42512,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub42512.stripped.N4.nii.gz
+sub-CC620557,F,74.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620557.stripped.N4.nii.gz
+sub-CC120469,F,25.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120469.stripped.N4.nii.gz
+ADNI399835,F,62.3,adni_storage/ADNI_nii_gz_bias_corrected/I399835_137_S_0994.stripped.N4.nii.gz
+ADNI379704,M,60.9,adni_storage/ADNI_nii_gz_bias_corrected/I379704_131_S_5138.stripped.N4.nii.gz
+DLBS0028574_session_1,M,50.35,dlbs_storage/DLBS_bias_corrected/0028574_session_1.stripped.N4.nii.gz
+ADNI359049,M,58.9,adni_storage/ADNI_nii_gz_bias_corrected/I359049_073_S_5090.stripped.N4.nii.gz
+sub-CC222367,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222367.stripped.N4.nii.gz
+fcon1000Beijing_sub56659,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub56659.stripped.N4.nii.gz
+fcon1000Cambridge_sub63661,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub63661.stripped.N4.nii.gz
+DLBS0028566_session_1,F,57.36,dlbs_storage/DLBS_bias_corrected/0028566_session_1.stripped.N4.nii.gz
+ADNI1079630,F,57.0,adni_storage/ADNI_nii_gz_bias_corrected/I1079630_082_S_6629.stripped.N4.nii.gz
+ADNI358859,F,58.6,adni_storage/ADNI_nii_gz_bias_corrected/I358859_130_S_4982.stripped.N4.nii.gz
+DLBS0028428_session_1,F,45.85,dlbs_storage/DLBS_bias_corrected/0028428_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub99085,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub99085.stripped.N4.nii.gz
+fcon1000Beijing_sub66528,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub66528.stripped.N4.nii.gz
+ADNI1332718,F,56.0,adni_storage/ADNI_nii_gz_bias_corrected/I1332718_123_S_6884.stripped.N4.nii.gz
+ADNI1031290,F,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I1031290_070_S_6542.stripped.N4.nii.gz
+fcon1000NewYork_sub22349,M,26.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub22349.stripped.N4.nii.gz
+fcon1000Cambridge_sub06987,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub06987.stripped.N4.nii.gz
+fcon1000Beijing_sub58029,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub58029.stripped.N4.nii.gz
+sub-CC610462,F,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610462.stripped.N4.nii.gz
+ADNI1029048,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I1029048_177_S_6409.stripped.N4.nii.gz
+DLBS0028502_session_1,F,82.54,dlbs_storage/DLBS_bias_corrected/0028502_session_1.stripped.N4.nii.gz
+DLBS0028514_session_1,M,34.44,dlbs_storage/DLBS_bias_corrected/0028514_session_1.stripped.N4.nii.gz
+fcon1000Cleveland_sub34189,M,55.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub34189.stripped.N4.nii.gz
+sub-CC520517,M,65.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520517.stripped.N4.nii.gz
+fcon1000Cambridge_sub68425,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub68425.stripped.N4.nii.gz
+fcon1000Beijing_sub42555,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub42555.stripped.N4.nii.gz
+ADNI10453853,M,61.0,adni_storage/ADNI_nii_gz_bias_corrected/I10453853_019_S_10102.stripped.N4.nii.gz
+fcon1000Beijing_sub30988,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub30988.stripped.N4.nii.gz
+fcon1000Beijing_sub42843,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub42843.stripped.N4.nii.gz
+fcon1000Berlin_sub27797,M,31.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub27797.stripped.N4.nii.gz
+sub-CC610099,F,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610099.stripped.N4.nii.gz
+fcon1000Cambridge_sub37374,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub37374.stripped.N4.nii.gz
+fcon1000Cambridge_sub34586,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub34586.stripped.N4.nii.gz
+fcon1000Cambridge_sub39142,M,25.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub39142.stripped.N4.nii.gz
+ADNI1559411,F,60.2,adni_storage/ADNI_nii_gz_bias_corrected/I1559411_082_S_6629.stripped.N4.nii.gz
+fcon1000Cleveland_sub58811,M,36.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub58811.stripped.N4.nii.gz
+fcon1000Cambridge_sub25058,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub25058.stripped.N4.nii.gz
+fcon1000Cambridge_sub90059,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub90059.stripped.N4.nii.gz
+fcon1000Beijing_sub88306,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub88306.stripped.N4.nii.gz
+ADNI386097,F,58.8,adni_storage/ADNI_nii_gz_bias_corrected/I386097_041_S_4874.stripped.N4.nii.gz
+DLBS0028382_session_1,F,88.07,dlbs_storage/DLBS_bias_corrected/0028382_session_1.stripped.N4.nii.gz
+sub-CC711245,M,85.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC711245.stripped.N4.nii.gz
+ADNI908586,M,63.0,adni_storage/ADNI_nii_gz_bias_corrected/I908586_012_S_6073.stripped.N4.nii.gz
+sub-CC221977,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221977.stripped.N4.nii.gz
+fcon1000NewYork_sub48803,M,37.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub48803.stripped.N4.nii.gz
+ADNI1002373,M,58.7,adni_storage/ADNI_nii_gz_bias_corrected/I1002373_014_S_6366.stripped.N4.nii.gz
+ADNI298226,F,64.1,adni_storage/ADNI_nii_gz_bias_corrected/I298226_023_S_4448.stripped.N4.nii.gz
+fcon1000Cleveland_sub46075,M,33.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub46075.stripped.N4.nii.gz
+fcon1000NewYork_sub26267,F,15.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub26267.stripped.N4.nii.gz
+fcon1000Cambridge_sub84064,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub84064.stripped.N4.nii.gz
+fcon1000Cleveland_sub67936,F,57.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub67936.stripped.N4.nii.gz
+sub-CC520065,M,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520065.stripped.N4.nii.gz
+sub-CC320417,M,39.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320417.stripped.N4.nii.gz
+ADNI20682,M,64.5,adni_storage/ADNI_nii_gz_bias_corrected/I20682_131_S_0691.stripped.N4.nii.gz
+sub-CC310473,M,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310473.stripped.N4.nii.gz
+fcon1000NewYork_sub15758,M,32.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub15758.stripped.N4.nii.gz
+fcon1000Berlin_sub47066,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub47066.stripped.N4.nii.gz
+fcon1000Beijing_sub35309,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub35309.stripped.N4.nii.gz
+fcon1000Beijing_sub02403,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub02403.stripped.N4.nii.gz
+fcon1000Cleveland_sub18566,F,33.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub18566.stripped.N4.nii.gz
+ADNI298228,F,64.1,adni_storage/ADNI_nii_gz_bias_corrected/I298228_023_S_4448.stripped.N4.nii.gz
+sub-CC420260,F,50.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420260.stripped.N4.nii.gz
+DLBS0028550_session_1,F,22.17,dlbs_storage/DLBS_bias_corrected/0028550_session_1.stripped.N4.nii.gz
+sub-CC710350,M,81.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710350.stripped.N4.nii.gz
+fcon1000AnnArbor_sub56028,F,26.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub56028.stripped.N4.nii.gz
+fcon1000Cambridge_sub61185,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub61185.stripped.N4.nii.gz
+ADNI1592044,F,57.3,adni_storage/ADNI_nii_gz_bias_corrected/I1592044_014_S_7080.stripped.N4.nii.gz
+DLBS0028469_session_1,M,50.46,dlbs_storage/DLBS_bias_corrected/0028469_session_1.stripped.N4.nii.gz
+sub-CC510355,F,65.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510355.stripped.N4.nii.gz
+fcon1000NewYork_sub24528,M,45.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub24528.stripped.N4.nii.gz
+DLBS0028486_session_1,F,37.81,dlbs_storage/DLBS_bias_corrected/0028486_session_1.stripped.N4.nii.gz
+fcon1000Bangor_sub27519,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub27519.stripped.N4.nii.gz
+DLBS0028609_session_1,M,85.77,dlbs_storage/DLBS_bias_corrected/0028609_session_1.stripped.N4.nii.gz
+fcon1000NewYork_sub44515,M,25.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub44515.stripped.N4.nii.gz
+ADNI10907107,F,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I10907107_073_S_10167.stripped.N4.nii.gz
+fcon1000Cambridge_sub77337,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub77337.stripped.N4.nii.gz
+sub-CC420061,M,57.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420061.stripped.N4.nii.gz
+fcon1000Cambridge_sub29800,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub29800.stripped.N4.nii.gz
+sub-CC710416,F,82.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710416.stripped.N4.nii.gz
+ADNI1321252,M,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I1321252_305_S_6744.stripped.N4.nii.gz
+fcon1000Cambridge_sub61436,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub61436.stripped.N4.nii.gz
+fcon1000AnnArbor_sub04619,M,18.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub04619.stripped.N4.nii.gz
+sub-CC420433,M,51.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420433.stripped.N4.nii.gz
+fcon1000Cambridge_sub99330,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub99330.stripped.N4.nii.gz
+sub-CC420454,M,54.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420454.stripped.N4.nii.gz
+sub-CC710494,M,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710494.stripped.N4.nii.gz
+fcon1000NewYork_sub54828,M,41.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub54828.stripped.N4.nii.gz
+ADNI339298,M,63.5,adni_storage/ADNI_nii_gz_bias_corrected/I339298_073_S_4552.stripped.N4.nii.gz
+ADNI271074,M,62.6,adni_storage/ADNI_nii_gz_bias_corrected/I271074_116_S_4195.stripped.N4.nii.gz
+sub-CC720497,F,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720497.stripped.N4.nii.gz
+ADNI942628,F,62.8,adni_storage/ADNI_nii_gz_bias_corrected/I942628_036_S_6134.stripped.N4.nii.gz
+ADNI372468,F,63.5,adni_storage/ADNI_nii_gz_bias_corrected/I372468_019_S_5019.stripped.N4.nii.gz
+fcon1000Newark_sub53422,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub53422.stripped.N4.nii.gz
+sub-CC510226,M,67.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510226.stripped.N4.nii.gz
+sub-CC420241,M,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420241.stripped.N4.nii.gz
+ADNI10417928,F,56.2,adni_storage/ADNI_nii_gz_bias_corrected/I10417928_037_S_10061.stripped.N4.nii.gz
+fcon1000Beijing_sub08455,M,25.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub08455.stripped.N4.nii.gz
+sub-CC510534,M,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510534.stripped.N4.nii.gz
+fcon1000Cambridge_sub61209,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub61209.stripped.N4.nii.gz
+sub-CC721434,M,84.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721434.stripped.N4.nii.gz
+sub-CC720023,F,82.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720023.stripped.N4.nii.gz
+DLBS0028442_session_1,F,29.32,dlbs_storage/DLBS_bias_corrected/0028442_session_1.stripped.N4.nii.gz
+sub-CC410220,F,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410220.stripped.N4.nii.gz
+fcon1000Baltimore_sub31837,F,34.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub31837.stripped.N4.nii.gz
+sub-CC320680,M,41.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320680.stripped.N4.nii.gz
+DLBS0028399_session_1,F,70.68,dlbs_storage/DLBS_bias_corrected/0028399_session_1.stripped.N4.nii.gz
+ADNI1527249,F,55.8,adni_storage/ADNI_nii_gz_bias_corrected/I1527249_035_S_7030.stripped.N4.nii.gz
+sub-CC410226,M,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410226.stripped.N4.nii.gz
+sub-CC620436,M,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620436.stripped.N4.nii.gz
+fcon1000Atlanta_sub61442,F,29.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub61442.stripped.N4.nii.gz
+sub-CC610288,M,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610288.stripped.N4.nii.gz
+fcon1000Beijing_sub87089,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub87089.stripped.N4.nii.gz
+fcon1000Beijing_sub53572,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub53572.stripped.N4.nii.gz
+DLBS0028346_session_1,F,57.48,dlbs_storage/DLBS_bias_corrected/0028346_session_1.stripped.N4.nii.gz
+ADNI1463003,M,57.1,adni_storage/ADNI_nii_gz_bias_corrected/I1463003_033_S_6976.stripped.N4.nii.gz
+fcon1000Newark_sub54933,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub54933.stripped.N4.nii.gz
+sub-CC120123,F,19.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120123.stripped.N4.nii.gz
+fcon1000Beijing_sub91952,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub91952.stripped.N4.nii.gz
+DLBS0028477_session_1,F,60.81,dlbs_storage/DLBS_bias_corrected/0028477_session_1.stripped.N4.nii.gz
+DLBS0028372_session_1,F,74.83,dlbs_storage/DLBS_bias_corrected/0028372_session_1.stripped.N4.nii.gz
+sub-CC610051,M,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610051.stripped.N4.nii.gz
+fcon1000Munchen_sub66933,M,72.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub66933.stripped.N4.nii.gz
+fcon1000Atlanta_sub61902,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub61902.stripped.N4.nii.gz
+sub-CC620496,F,70.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620496.stripped.N4.nii.gz
+fcon1000Beijing_sub10973,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub10973.stripped.N4.nii.gz
+fcon1000NewYork_sub21212,M,30.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub21212.stripped.N4.nii.gz
+sub-CC520055,F,61.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520055.stripped.N4.nii.gz
+ADNI1004030,F,64.3,adni_storage/ADNI_nii_gz_bias_corrected/I1004030_023_S_6400.stripped.N4.nii.gz
+ADNI925573,M,63.4,adni_storage/ADNI_nii_gz_bias_corrected/I925573_168_S_6098.stripped.N4.nii.gz
+fcon1000Baltimore_sub81887,M,32.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub81887.stripped.N4.nii.gz
+fcon1000NewYork_sub77203,M,40.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub77203.stripped.N4.nii.gz
+sub-CC410177,M,57.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410177.stripped.N4.nii.gz
+DLBS0028338_session_1,F,30.26,dlbs_storage/DLBS_bias_corrected/0028338_session_1.stripped.N4.nii.gz
+ADNI10666918,F,59.7,adni_storage/ADNI_nii_gz_bias_corrected/I10666918_301_S_10139.stripped.N4.nii.gz
+fcon1000Cambridge_sub09015,F,25.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub09015.stripped.N4.nii.gz
+DLBS0028455_session_1,F,72.56,dlbs_storage/DLBS_bias_corrected/0028455_session_1.stripped.N4.nii.gz
+sub-CC420197,F,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420197.stripped.N4.nii.gz
+sub-CC710313,M,81.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710313.stripped.N4.nii.gz
+fcon1000Atlanta_sub55652,F,25.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub55652.stripped.N4.nii.gz
+fcon1000Leiden_sub10481,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub10481.stripped.N4.nii.gz
+fcon1000Cambridge_sub64308,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub64308.stripped.N4.nii.gz
+fcon1000Beijing_sub10869,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub10869.stripped.N4.nii.gz
+sub-CC520247,M,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520247.stripped.N4.nii.gz
+DLBS0028367_session_1,F,29.58,dlbs_storage/DLBS_bias_corrected/0028367_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub35512,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub35512.stripped.N4.nii.gz
+ADNI342679,F,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I342679_073_S_5016.stripped.N4.nii.gz
+ADNI1215081,F,56.9,adni_storage/ADNI_nii_gz_bias_corrected/I1215081_116_S_6550.stripped.N4.nii.gz
+sub-CC510304,F,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510304.stripped.N4.nii.gz
+sub-CC620549,F,73.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620549.stripped.N4.nii.gz
+DLBS0028569_session_1,M,67.65,dlbs_storage/DLBS_bias_corrected/0028569_session_1.stripped.N4.nii.gz
+sub-CC320576,F,45.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320576.stripped.N4.nii.gz
+fcon1000Cleveland_sub20003,M,39.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub20003.stripped.N4.nii.gz
+sub-CC220115,F,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220115.stripped.N4.nii.gz
+ADNI883189,F,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I883189_141_S_6042.stripped.N4.nii.gz
+ADNI415587,M,57.2,adni_storage/ADNI_nii_gz_bias_corrected/I415587_009_S_5252.stripped.N4.nii.gz
+fcon1000Atlanta_sub18702,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub18702.stripped.N4.nii.gz
+sub-CC723395,F,86.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC723395.stripped.N4.nii.gz
+fcon1000NewYork_sub53710,F,34.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub53710.stripped.N4.nii.gz
+ADNI372477,F,63.5,adni_storage/ADNI_nii_gz_bias_corrected/I372477_019_S_5019.stripped.N4.nii.gz
+ADNI238752,M,56.0,adni_storage/ADNI_nii_gz_bias_corrected/I238752_014_S_4039.stripped.N4.nii.gz
+sub-CC221031,M,38.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221031.stripped.N4.nii.gz
+fcon1000Cambridge_sub51172,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub51172.stripped.N4.nii.gz
+sub-CC620515,F,75.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620515.stripped.N4.nii.gz
+sub-CC320267,M,38.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320267.stripped.N4.nii.gz
+sub-CC221038,M,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221038.stripped.N4.nii.gz
+fcon1000AnnArbor_sub42616,F,21.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub42616.stripped.N4.nii.gz
+sub-CC321595,F,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321595.stripped.N4.nii.gz
+DLBS0028496_session_1,M,22.44,dlbs_storage/DLBS_bias_corrected/0028496_session_1.stripped.N4.nii.gz
+sub-CC720290,M,84.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720290.stripped.N4.nii.gz
+sub-CC520168,M,59.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520168.stripped.N4.nii.gz
+fcon1000Berlin_sub97162,M,28.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub97162.stripped.N4.nii.gz
+DLBS0028541_session_1,F,45.8,dlbs_storage/DLBS_bias_corrected/0028541_session_1.stripped.N4.nii.gz
+DLBS0028402_session_1,M,86.41,dlbs_storage/DLBS_bias_corrected/0028402_session_1.stripped.N4.nii.gz
+fcon1000Milwaukee_sub45019,F,44.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub45019.stripped.N4.nii.gz
+ADNI46723,M,62.9,adni_storage/ADNI_nii_gz_bias_corrected/I46723_016_S_0769.stripped.N4.nii.gz
+sub-CC720986,M,83.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720986.stripped.N4.nii.gz
+sub-CC710342,F,83.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710342.stripped.N4.nii.gz
+sub-CC520127,M,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520127.stripped.N4.nii.gz
+ADNI117520,F,61.7,adni_storage/ADNI_nii_gz_bias_corrected/I117520_051_S_1072.stripped.N4.nii.gz
+fcon1000Beijing_sub33991,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub33991.stripped.N4.nii.gz
+sub-CC420180,F,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420180.stripped.N4.nii.gz
+sub-CC610022,F,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610022.stripped.N4.nii.gz
+sub-CC521040,F,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC521040.stripped.N4.nii.gz
+fcon1000Beijing_sub62083,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub62083.stripped.N4.nii.gz
+fcon1000Atlanta_sub26938,M,31.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub26938.stripped.N4.nii.gz
+ADNI958917,F,55.6,adni_storage/ADNI_nii_gz_bias_corrected/I958917_070_S_6191.stripped.N4.nii.gz
+sub-CC420623,M,51.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420623.stripped.N4.nii.gz
+fcon1000NewYork_sub45217,M,40.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub45217.stripped.N4.nii.gz
+sub-CC310135,F,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310135.stripped.N4.nii.gz
+ADNI1427086,F,60.2,adni_storage/ADNI_nii_gz_bias_corrected/I1427086_003_S_6924.stripped.N4.nii.gz
+DLBS0028456_session_1,M,53.75,dlbs_storage/DLBS_bias_corrected/0028456_session_1.stripped.N4.nii.gz
+ADNI1549623,F,62.4,adni_storage/ADNI_nii_gz_bias_corrected/I1549623_035_S_7043.stripped.N4.nii.gz
+ADNI1485047,F,58.9,adni_storage/ADNI_nii_gz_bias_corrected/I1485047_116_S_6550.stripped.N4.nii.gz
+fcon1000Berlin_sub75506,M,28.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub75506.stripped.N4.nii.gz
+fcon1000AnnArbor_sub47659,M,33.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub47659.stripped.N4.nii.gz
+fcon1000NewYork_sub12486,M,40.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub12486.stripped.N4.nii.gz
+DLBS0028364_session_1,F,24.49,dlbs_storage/DLBS_bias_corrected/0028364_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub98624,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub98624.stripped.N4.nii.gz
+ADNI982856,F,62.6,adni_storage/ADNI_nii_gz_bias_corrected/I982856_109_S_6218.stripped.N4.nii.gz
+sub-CC321899,F,49.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321899.stripped.N4.nii.gz
+sub-CC310397,F,41.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310397.stripped.N4.nii.gz
+fcon1000NewYork_sub51677,F,25.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub51677.stripped.N4.nii.gz
+fcon1000AnnArbor_sub72215,M,20.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub72215.stripped.N4.nii.gz
+sub-CC221107,M,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221107.stripped.N4.nii.gz
+fcon1000Newark_sub76647,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub76647.stripped.N4.nii.gz
+fcon1000Atlanta_sub75153,F,32.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub75153.stripped.N4.nii.gz
+fcon1000Cambridge_sub17737,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub17737.stripped.N4.nii.gz
+ADNI296138,M,63.9,adni_storage/ADNI_nii_gz_bias_corrected/I296138_116_S_4625.stripped.N4.nii.gz
+sub-CC410094,F,54.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410094.stripped.N4.nii.gz
+fcon1000Cambridge_sub41567,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub41567.stripped.N4.nii.gz
+fcon1000Atlanta_sub24972,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub24972.stripped.N4.nii.gz
+fcon1000AnnArbor_sub90950,M,66.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub90950.stripped.N4.nii.gz
+fcon1000Beijing_sub35806,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub35806.stripped.N4.nii.gz
+sub-CC721532,F,87.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721532.stripped.N4.nii.gz
+fcon1000Cambridge_sub51512,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub51512.stripped.N4.nii.gz
+ADNI946333,F,58.4,adni_storage/ADNI_nii_gz_bias_corrected/I946333_024_S_6033.stripped.N4.nii.gz
+fcon1000Cambridge_sub71849,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub71849.stripped.N4.nii.gz
+ADNI23823,M,62.3,adni_storage/ADNI_nii_gz_bias_corrected/I23823_016_S_0769.stripped.N4.nii.gz
+DLBS0028462_session_1,M,31.03,dlbs_storage/DLBS_bias_corrected/0028462_session_1.stripped.N4.nii.gz
+fcon1000Milwaukee_sub17004,M,57.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub17004.stripped.N4.nii.gz
+ADNI10460262,F,56.4,adni_storage/ADNI_nii_gz_bias_corrected/I10460262_011_S_10041.stripped.N4.nii.gz
+sub-CC510208,F,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510208.stripped.N4.nii.gz
+DLBS0028524_session_1,F,56.55,dlbs_storage/DLBS_bias_corrected/0028524_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub73812,M,28.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub73812.stripped.N4.nii.gz
+sub-CC410284,F,50.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410284.stripped.N4.nii.gz
+fcon1000Atlanta_sub58250,F,54.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub58250.stripped.N4.nii.gz
+fcon1000Newark_sub71743,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub71743.stripped.N4.nii.gz
+ADNI115774,F,59.2,adni_storage/ADNI_nii_gz_bias_corrected/I115774_033_S_0733.stripped.N4.nii.gz
+ADNI394452,M,62.3,adni_storage/ADNI_nii_gz_bias_corrected/I394452_135_S_4954.stripped.N4.nii.gz
+ADNI358858,F,58.6,adni_storage/ADNI_nii_gz_bias_corrected/I358858_130_S_4982.stripped.N4.nii.gz
+sub-CC321203,M,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321203.stripped.N4.nii.gz
+fcon1000Cambridge_sub94304,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub94304.stripped.N4.nii.gz
+ADNI1591453,F,62.7,adni_storage/ADNI_nii_gz_bias_corrected/I1591453_014_S_7072.stripped.N4.nii.gz
+DLBS0028526_session_1,M,42.6,dlbs_storage/DLBS_bias_corrected/0028526_session_1.stripped.N4.nii.gz
+sub-CC310400,M,43.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310400.stripped.N4.nii.gz
+fcon1000Bangor_sub87568,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub87568.stripped.N4.nii.gz
+sub-CC520200,F,67.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520200.stripped.N4.nii.gz
+sub-CC722891,F,84.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC722891.stripped.N4.nii.gz
+sub-CC420100,M,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420100.stripped.N4.nii.gz
+sub-CC520053,M,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520053.stripped.N4.nii.gz
+fcon1000Berlin_sub23506,F,27.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub23506.stripped.N4.nii.gz
+sub-CC520287,F,59.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520287.stripped.N4.nii.gz
+DLBS0028568_session_1,M,41.17,dlbs_storage/DLBS_bias_corrected/0028568_session_1.stripped.N4.nii.gz
+ADNI1032531,F,55.9,adni_storage/ADNI_nii_gz_bias_corrected/I1032531_116_S_6550.stripped.N4.nii.gz
+sub-CC320621,F,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320621.stripped.N4.nii.gz
+fcon1000Cambridge_sub62424,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub62424.stripped.N4.nii.gz
+fcon1000Cambridge_sub43304,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub43304.stripped.N4.nii.gz
+sub-CC610061,F,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610061.stripped.N4.nii.gz
+sub-CC320478,M,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320478.stripped.N4.nii.gz
+sub-CC722421,F,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC722421.stripped.N4.nii.gz
+ADNI1469780,M,62.4,adni_storage/ADNI_nii_gz_bias_corrected/I1469780_073_S_6669.stripped.N4.nii.gz
+fcon1000Munchen_sub64202,F,67.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub64202.stripped.N4.nii.gz
+fcon1000Beijing_sub33747,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub33747.stripped.N4.nii.gz
+ADNI37255,M,62.2,adni_storage/ADNI_nii_gz_bias_corrected/I37255_067_S_1185.stripped.N4.nii.gz
+sub-CC420229,M,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420229.stripped.N4.nii.gz
+sub-CC221002,F,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221002.stripped.N4.nii.gz
+DLBS0028599_session_1,F,62.14,dlbs_storage/DLBS_bias_corrected/0028599_session_1.stripped.N4.nii.gz
+sub-CC320888,F,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320888.stripped.N4.nii.gz
+DLBS0028618_session_1,F,26.36,dlbs_storage/DLBS_bias_corrected/0028618_session_1.stripped.N4.nii.gz
+sub-CC610614,F,74.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610614.stripped.N4.nii.gz
+sub-CC710679,M,84.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710679.stripped.N4.nii.gz
+sub-CC222125,M,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222125.stripped.N4.nii.gz
+ADNI10658149,F,56.5,adni_storage/ADNI_nii_gz_bias_corrected/I10658149_016_S_10124.stripped.N4.nii.gz
+sub-CC720622,F,81.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720622.stripped.N4.nii.gz
+sub-CC712027,M,87.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC712027.stripped.N4.nii.gz
+sub-CC221487,M,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221487.stripped.N4.nii.gz
+sub-CC220352,F,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220352.stripped.N4.nii.gz
+ADNI46724,M,62.9,adni_storage/ADNI_nii_gz_bias_corrected/I46724_016_S_0769.stripped.N4.nii.gz
+fcon1000Berlin_sub57028,F,37.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub57028.stripped.N4.nii.gz
+fcon1000Beijing_sub06899,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub06899.stripped.N4.nii.gz
+fcon1000Cambridge_sub00294,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub00294.stripped.N4.nii.gz
+fcon1000Leiden_sub40907,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub40907.stripped.N4.nii.gz
+DLBS0028355_session_1,M,70.7,dlbs_storage/DLBS_bias_corrected/0028355_session_1.stripped.N4.nii.gz
+ADNI1476165,M,64.3,adni_storage/ADNI_nii_gz_bias_corrected/I1476165_305_S_6744.stripped.N4.nii.gz
+ADNI1484638,F,62.1,adni_storage/ADNI_nii_gz_bias_corrected/I1484638_067_S_6986.stripped.N4.nii.gz
+fcon1000Beijing_sub26713,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub26713.stripped.N4.nii.gz
+DLBS0028510_session_1,F,51.3,dlbs_storage/DLBS_bias_corrected/0028510_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub20563,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub20563.stripped.N4.nii.gz
+fcon1000Newark_sub36023,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub36023.stripped.N4.nii.gz
+fcon1000Berlin_sub27536,M,25.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub27536.stripped.N4.nii.gz
+fcon1000Milwaukee_sub75919,F,50.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub75919.stripped.N4.nii.gz
+sub-CC110062,M,20.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110062.stripped.N4.nii.gz
+ADNI347408,F,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I347408_130_S_4997.stripped.N4.nii.gz
+fcon1000Leiden_sub18456,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub18456.stripped.N4.nii.gz
+ADNI1571742,F,58.6,adni_storage/ADNI_nii_gz_bias_corrected/I1571742_067_S_7061.stripped.N4.nii.gz
+sub-CC610469,F,73.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610469.stripped.N4.nii.gz
+fcon1000Munchen_sub28902,M,65.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub28902.stripped.N4.nii.gz
+ADNI67599,F,58.2,adni_storage/ADNI_nii_gz_bias_corrected/I67599_033_S_0733.stripped.N4.nii.gz
+sub-CC320686,F,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320686.stripped.N4.nii.gz
+sub-CC220098,F,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220098.stripped.N4.nii.gz
+DLBS0028489_session_1,F,70.28,dlbs_storage/DLBS_bias_corrected/0028489_session_1.stripped.N4.nii.gz
+fcon1000Leiden_sub66131,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub66131.stripped.N4.nii.gz
+DLBS0028626_session_1,M,48.88,dlbs_storage/DLBS_bias_corrected/0028626_session_1.stripped.N4.nii.gz
+sub-CC720511,M,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720511.stripped.N4.nii.gz
+fcon1000Beijing_sub28403,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub28403.stripped.N4.nii.gz
+ADNI1478584,F,57.4,adni_storage/ADNI_nii_gz_bias_corrected/I1478584_037_S_6963.stripped.N4.nii.gz
+sub-CC721648,F,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721648.stripped.N4.nii.gz
+ADNI267802,F,64.1,adni_storage/ADNI_nii_gz_bias_corrected/I267802_037_S_4028.stripped.N4.nii.gz
+fcon1000Beijing_sub55856,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub55856.stripped.N4.nii.gz
+fcon1000NewYork_sub10582,M,47.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub10582.stripped.N4.nii.gz
+ADNI87683,F,57.8,adni_storage/ADNI_nii_gz_bias_corrected/I87683_133_S_0913.stripped.N4.nii.gz
+ADNI883190,F,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I883190_141_S_6042.stripped.N4.nii.gz
+fcon1000Milwaukee_sub51182,M,57.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub51182.stripped.N4.nii.gz
+sub-CC720188,M,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720188.stripped.N4.nii.gz
+sub-CC420231,M,54.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420231.stripped.N4.nii.gz
+ADNI382181,F,62.8,adni_storage/ADNI_nii_gz_bias_corrected/I382181_018_S_5240.stripped.N4.nii.gz
+DLBS0028501_session_1,F,88.36,dlbs_storage/DLBS_bias_corrected/0028501_session_1.stripped.N4.nii.gz
+DLBS0028584_session_1,F,51.24,dlbs_storage/DLBS_bias_corrected/0028584_session_1.stripped.N4.nii.gz
+ADNI336509,M,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I336509_135_S_4954.stripped.N4.nii.gz
+sub-CC610039,M,70.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610039.stripped.N4.nii.gz
+fcon1000Cambridge_sub80557,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub80557.stripped.N4.nii.gz
+fcon1000Beijing_sub77440,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub77440.stripped.N4.nii.gz
+DLBS0028532_session_1,F,34.11,dlbs_storage/DLBS_bias_corrected/0028532_session_1.stripped.N4.nii.gz
+DLBS0028505_session_1,M,20.75,dlbs_storage/DLBS_bias_corrected/0028505_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub45604,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub45604.stripped.N4.nii.gz
+sub-CC610625,F,70.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610625.stripped.N4.nii.gz
+ADNI10883387,M,60.2,adni_storage/ADNI_nii_gz_bias_corrected/I10883387_033_S_6976.stripped.N4.nii.gz
+sub-CC420435,F,53.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420435.stripped.N4.nii.gz
+fcon1000Milwaukee_sub84314,M,58.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub84314.stripped.N4.nii.gz
+DLBS0028559_session_1,F,25.01,dlbs_storage/DLBS_bias_corrected/0028559_session_1.stripped.N4.nii.gz
+fcon1000NewYork_sub27123,M,23.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub27123.stripped.N4.nii.gz
+fcon1000Beijing_sub32587,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub32587.stripped.N4.nii.gz
+DLBS0028343_session_1,F,61.27,dlbs_storage/DLBS_bias_corrected/0028343_session_1.stripped.N4.nii.gz
+sub-CC720516,F,84.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720516.stripped.N4.nii.gz
+sub-CC510086,M,65.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510086.stripped.N4.nii.gz
+fcon1000Cambridge_sub84504,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub84504.stripped.N4.nii.gz
+fcon1000Beijing_sub19642,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub19642.stripped.N4.nii.gz
+ADNI369303,F,58.8,adni_storage/ADNI_nii_gz_bias_corrected/I369303_130_S_4982.stripped.N4.nii.gz
+fcon1000Cleveland_sub17946,F,29.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub17946.stripped.N4.nii.gz
+sub-CC222555,M,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222555.stripped.N4.nii.gz
+ADNI10972872,F,64.6,adni_storage/ADNI_nii_gz_bias_corrected/I10972872_033_S_10005.stripped.N4.nii.gz
+sub-CC412004,M,54.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC412004.stripped.N4.nii.gz
+DLBS0028517_session_1,M,23.14,dlbs_storage/DLBS_bias_corrected/0028517_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub80163,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub80163.stripped.N4.nii.gz
+ADNI346548,F,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I346548_137_S_0994.stripped.N4.nii.gz
+sub-CC710664,F,82.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710664.stripped.N4.nii.gz
+fcon1000Milwaukee_sub36386,M,57.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub36386.stripped.N4.nii.gz
+fcon1000Cambridge_sub86637,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub86637.stripped.N4.nii.gz
+sub-CC420075,M,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420075.stripped.N4.nii.gz
+fcon1000Beijing_sub20246,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub20246.stripped.N4.nii.gz
+fcon1000Beijing_sub85030,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub85030.stripped.N4.nii.gz
+sub-CC320342,F,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320342.stripped.N4.nii.gz
+ADNI970943,F,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I970943_153_S_6237.stripped.N4.nii.gz
+sub-CC320759,F,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320759.stripped.N4.nii.gz
+ADNI10939738,F,62.5,adni_storage/ADNI_nii_gz_bias_corrected/I10939738_301_S_10230.stripped.N4.nii.gz
+sub-CC420091,M,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420091.stripped.N4.nii.gz
+DLBS0028611_session_1,F,46.79,dlbs_storage/DLBS_bias_corrected/0028611_session_1.stripped.N4.nii.gz
+sub-CC720180,M,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720180.stripped.N4.nii.gz
+sub-CC610496,M,70.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610496.stripped.N4.nii.gz
+sub-CC410119,M,57.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410119.stripped.N4.nii.gz
+fcon1000Beijing_sub92544,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub92544.stripped.N4.nii.gz
+sub-CC610058,M,73.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610058.stripped.N4.nii.gz
+fcon1000Bangor_sub73082,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub73082.stripped.N4.nii.gz
+sub-CC220635,F,36.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220635.stripped.N4.nii.gz
+sub-CC120987,F,20.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120987.stripped.N4.nii.gz
+sub-CC210172,F,31.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210172.stripped.N4.nii.gz
+fcon1000Milwaukee_sub56108,F,51.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub56108.stripped.N4.nii.gz
+DLBS0028330_session_1,F,35.36,dlbs_storage/DLBS_bias_corrected/0028330_session_1.stripped.N4.nii.gz
+ADNI1002815,F,64.2,adni_storage/ADNI_nii_gz_bias_corrected/I1002815_023_S_6374.stripped.N4.nii.gz
+sub-CC420137,M,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420137.stripped.N4.nii.gz
+DLBS0028504_session_1,F,73.07,dlbs_storage/DLBS_bias_corrected/0028504_session_1.stripped.N4.nii.gz
+sub-CC620114,M,73.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620114.stripped.N4.nii.gz
+DLBS0028591_session_1,M,86.25,dlbs_storage/DLBS_bias_corrected/0028591_session_1.stripped.N4.nii.gz
+fcon1000Leiden_sub68050,M,25.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub68050.stripped.N4.nii.gz
+DLBS0028395_session_1,M,83.87,dlbs_storage/DLBS_bias_corrected/0028395_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub13959,M,31.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub13959.stripped.N4.nii.gz
+fcon1000Cambridge_sub34741,F,27.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub34741.stripped.N4.nii.gz
+DLBS0028616_session_1,F,61.54,dlbs_storage/DLBS_bias_corrected/0028616_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub46541,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub46541.stripped.N4.nii.gz
+sub-CC120049,M,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120049.stripped.N4.nii.gz
+sub-CC610046,M,72.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610046.stripped.N4.nii.gz
+DLBS0028418_session_1,M,70.49,dlbs_storage/DLBS_bias_corrected/0028418_session_1.stripped.N4.nii.gz
+sub-CC721618,F,81.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721618.stripped.N4.nii.gz
+fcon1000Beijing_sub92490,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub92490.stripped.N4.nii.gz
+fcon1000Leiden_sub08518,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub08518.stripped.N4.nii.gz
+DLBS0028416_session_1,M,31.58,dlbs_storage/DLBS_bias_corrected/0028416_session_1.stripped.N4.nii.gz
+ADNI10883427,F,62.7,adni_storage/ADNI_nii_gz_bias_corrected/I10883427_016_S_10138.stripped.N4.nii.gz
+sub-CC121795,F,25.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121795.stripped.N4.nii.gz
+DLBS0028431_session_1,M,47.63,dlbs_storage/DLBS_bias_corrected/0028431_session_1.stripped.N4.nii.gz
+ADNI1400867,F,58.7,adni_storage/ADNI_nii_gz_bias_corrected/I1400867_019_S_6635.stripped.N4.nii.gz
+DLBS0028556_session_1,F,36.97,dlbs_storage/DLBS_bias_corrected/0028556_session_1.stripped.N4.nii.gz
+ADNI67598,F,58.2,adni_storage/ADNI_nii_gz_bias_corrected/I67598_033_S_0733.stripped.N4.nii.gz
+sub-CC321137,F,41.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321137.stripped.N4.nii.gz
+fcon1000Cambridge_sub53107,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub53107.stripped.N4.nii.gz
+fcon1000Beijing_sub98353,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub98353.stripped.N4.nii.gz
+sub-CC710131,M,85.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710131.stripped.N4.nii.gz
+DLBS0028564_session_1,F,40.21,dlbs_storage/DLBS_bias_corrected/0028564_session_1.stripped.N4.nii.gz
+ADNI86564,F,57.3,adni_storage/ADNI_nii_gz_bias_corrected/I86564_016_S_1121.stripped.N4.nii.gz
+DLBS0028461_session_1,F,53.95,dlbs_storage/DLBS_bias_corrected/0028461_session_1.stripped.N4.nii.gz
+ADNI70746,M,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I70746_016_S_0769.stripped.N4.nii.gz
+sub-CC320059,M,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320059.stripped.N4.nii.gz
+ADNI1433167,F,61.7,adni_storage/ADNI_nii_gz_bias_corrected/I1433167_024_S_6033.stripped.N4.nii.gz
+ADNI487386,M,61.1,adni_storage/ADNI_nii_gz_bias_corrected/I487386_128_S_4599.stripped.N4.nii.gz
+DLBS0028511_session_1,F,23.55,dlbs_storage/DLBS_bias_corrected/0028511_session_1.stripped.N4.nii.gz
+sub-CC610575,F,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610575.stripped.N4.nii.gz
+sub-CC610227,F,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610227.stripped.N4.nii.gz
+ADNI1031289,F,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I1031289_070_S_6542.stripped.N4.nii.gz
+fcon1000NewYork_sub63915,M,44.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub63915.stripped.N4.nii.gz
+sub-CC220828,F,33.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220828.stripped.N4.nii.gz
+DLBS0028551_session_1,M,24.66,dlbs_storage/DLBS_bias_corrected/0028551_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub08947,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub08947.stripped.N4.nii.gz
+sub-CC610050,F,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610050.stripped.N4.nii.gz
+DLBS0028433_session_1,F,80.5,dlbs_storage/DLBS_bias_corrected/0028433_session_1.stripped.N4.nii.gz
+ADNI1282405,F,57.7,adni_storage/ADNI_nii_gz_bias_corrected/I1282405_019_S_6635.stripped.N4.nii.gz
+sub-CC120816,M,23.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120816.stripped.N4.nii.gz
+sub-CC120276,F,23.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120276.stripped.N4.nii.gz
+ADNI65945,M,62.7,adni_storage/ADNI_nii_gz_bias_corrected/I65945_067_S_1185.stripped.N4.nii.gz
+fcon1000Beijing_sub55301,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub55301.stripped.N4.nii.gz
+ADNI839474,F,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I839474_141_S_6008.stripped.N4.nii.gz
+fcon1000Cambridge_sub73477,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub73477.stripped.N4.nii.gz
+fcon1000Beijing_sub12220,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub12220.stripped.N4.nii.gz
+sub-CC210422,M,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210422.stripped.N4.nii.gz
+ADNI10435128,F,55.5,adni_storage/ADNI_nii_gz_bias_corrected/I10435128_035_S_10093.stripped.N4.nii.gz
+fcon1000Cambridge_sub26348,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub26348.stripped.N4.nii.gz
+fcon1000Beijing_sub10277,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub10277.stripped.N4.nii.gz
+DLBS0028329_session_1,M,30.74,dlbs_storage/DLBS_bias_corrected/0028329_session_1.stripped.N4.nii.gz
+ADNI23822,M,62.3,adni_storage/ADNI_nii_gz_bias_corrected/I23822_016_S_0769.stripped.N4.nii.gz
+ADNI1184047,F,62.6,adni_storage/ADNI_nii_gz_bias_corrected/I1184047_022_S_6013.stripped.N4.nii.gz
+sub-CC220567,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220567.stripped.N4.nii.gz
+sub-CC120309,M,27.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120309.stripped.N4.nii.gz
+fcon1000Cambridge_sub43358,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub43358.stripped.N4.nii.gz
+fcon1000Cleveland_sub46739,F,60.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub46739.stripped.N4.nii.gz
+sub-CC212153,F,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC212153.stripped.N4.nii.gz
+ADNI1017005,M,59.0,adni_storage/ADNI_nii_gz_bias_corrected/I1017005_116_S_6428.stripped.N4.nii.gz
+fcon1000Baltimore_sub17017,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub17017.stripped.N4.nii.gz
+DLBS0028449_session_1,F,28.96,dlbs_storage/DLBS_bias_corrected/0028449_session_1.stripped.N4.nii.gz
+ADNI1536458,F,61.6,adni_storage/ADNI_nii_gz_bias_corrected/I1536458_016_S_7039.stripped.N4.nii.gz
+sub-CC320297,M,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320297.stripped.N4.nii.gz
+ADNI1156695,F,64.0,adni_storage/ADNI_nii_gz_bias_corrected/I1156695_023_S_6723.stripped.N4.nii.gz
+ADNI415207,F,64.0,adni_storage/ADNI_nii_gz_bias_corrected/I415207_019_S_5019.stripped.N4.nii.gz
+fcon1000Beijing_sub89238,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub89238.stripped.N4.nii.gz
+ADNI10461419,F,58.2,adni_storage/ADNI_nii_gz_bias_corrected/I10461419_011_S_10044.stripped.N4.nii.gz
+sub-CC721292,M,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721292.stripped.N4.nii.gz
+ADNI495305,F,57.3,adni_storage/ADNI_nii_gz_bias_corrected/I495305_128_S_4607.stripped.N4.nii.gz
+sub-CC220323,F,33.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220323.stripped.N4.nii.gz
+ADNI346547,F,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I346547_137_S_0994.stripped.N4.nii.gz
+fcon1000Cambridge_sub81289,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub81289.stripped.N4.nii.gz
+ADNI10460278,F,61.6,adni_storage/ADNI_nii_gz_bias_corrected/I10460278_019_S_10106.stripped.N4.nii.gz
+fcon1000AnnArbor_sub45660,M,16.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub45660.stripped.N4.nii.gz
+sub-CC721114,F,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721114.stripped.N4.nii.gz
+fcon1000Beijing_sub95755,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub95755.stripped.N4.nii.gz
+sub-CC221980,M,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221980.stripped.N4.nii.gz
+ADNI461110,F,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I461110_137_S_0994.stripped.N4.nii.gz
+sub-CC620259,F,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620259.stripped.N4.nii.gz
+ADNI1070046,M,61.4,adni_storage/ADNI_nii_gz_bias_corrected/I1070046_006_S_6610.stripped.N4.nii.gz
+ADNI37605,F,60.1,adni_storage/ADNI_nii_gz_bias_corrected/I37605_051_S_1072.stripped.N4.nii.gz
+fcon1000Cleveland_sub82518,M,30.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub82518.stripped.N4.nii.gz
+ADNI1373547,F,62.3,adni_storage/ADNI_nii_gz_bias_corrected/I1373547_041_S_6801.stripped.N4.nii.gz
+ADNI1181047,M,60.0,adni_storage/ADNI_nii_gz_bias_corrected/I1181047_116_S_6428.stripped.N4.nii.gz
+fcon1000Leiden_sub97690,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub97690.stripped.N4.nii.gz
+fcon1000Beijing_sub58332,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub58332.stripped.N4.nii.gz
+sub-CC120462,F,18.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120462.stripped.N4.nii.gz
+fcon1000Milwaukee_sub93170,F,65.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub93170.stripped.N4.nii.gz
+fcon1000Berlin_sub47791,M,31.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub47791.stripped.N4.nii.gz
+DLBS0028387_session_1,F,54.68,dlbs_storage/DLBS_bias_corrected/0028387_session_1.stripped.N4.nii.gz
+sub-CC720400,F,86.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720400.stripped.N4.nii.gz
+DLBS0028411_session_1,F,83.09,dlbs_storage/DLBS_bias_corrected/0028411_session_1.stripped.N4.nii.gz
+ADNI77627,M,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I77627_020_S_1288.stripped.N4.nii.gz
+ADNI487400,M,61.1,adni_storage/ADNI_nii_gz_bias_corrected/I487400_128_S_4599.stripped.N4.nii.gz
+fcon1000NewYork_sub47633,F,37.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub47633.stripped.N4.nii.gz
+sub-CC410251,F,53.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410251.stripped.N4.nii.gz
+sub-CC410129,M,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410129.stripped.N4.nii.gz
+sub-CC720407,M,82.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720407.stripped.N4.nii.gz
+DLBS0028603_session_1,M,35.96,dlbs_storage/DLBS_bias_corrected/0028603_session_1.stripped.N4.nii.gz
+sub-CC410289,F,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410289.stripped.N4.nii.gz
+DLBS0028519_session_1,M,64.79,dlbs_storage/DLBS_bias_corrected/0028519_session_1.stripped.N4.nii.gz
+ADNI1331165,F,57.9,adni_storage/ADNI_nii_gz_bias_corrected/I1331165_116_S_6550.stripped.N4.nii.gz
+fcon1000AnnArbor_sub59573,F,19.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub59573.stripped.N4.nii.gz
+ADNI1028373,M,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I1028373_109_S_6406.stripped.N4.nii.gz
+sub-CC320687,F,41.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320687.stripped.N4.nii.gz
+fcon1000Cambridge_sub90681,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub90681.stripped.N4.nii.gz
+ADNI369298,F,58.8,adni_storage/ADNI_nii_gz_bias_corrected/I369298_130_S_4982.stripped.N4.nii.gz
+fcon1000NewYork_sub33062,F,34.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub33062.stripped.N4.nii.gz
+fcon1000AnnArbor_sub47191,M,27.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub47191.stripped.N4.nii.gz
+ADNI47972,F,62.8,adni_storage/ADNI_nii_gz_bias_corrected/I47972_067_S_1253.stripped.N4.nii.gz
+ADNI346266,F,61.0,adni_storage/ADNI_nii_gz_bias_corrected/I346266_036_S_4894.stripped.N4.nii.gz
+sub-CC420060,M,51.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420060.stripped.N4.nii.gz
+fcon1000Beijing_sub00440,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub00440.stripped.N4.nii.gz
+sub-CC520002,F,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520002.stripped.N4.nii.gz
+sub-CC710858,M,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710858.stripped.N4.nii.gz
+ADNI1002374,M,58.7,adni_storage/ADNI_nii_gz_bias_corrected/I1002374_014_S_6366.stripped.N4.nii.gz
+fcon1000Atlanta_sub86323,F,31.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub86323.stripped.N4.nii.gz
+fcon1000Baltimore_sub30072,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub30072.stripped.N4.nii.gz
+fcon1000Beijing_sub01018,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub01018.stripped.N4.nii.gz
+sub-CC120470,F,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120470.stripped.N4.nii.gz
+fcon1000Cleveland_sub75398,F,43.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub75398.stripped.N4.nii.gz
+sub-CC321107,M,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321107.stripped.N4.nii.gz
+sub-CC520253,M,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520253.stripped.N4.nii.gz
+DLBS0028509_session_1,F,51.39,dlbs_storage/DLBS_bias_corrected/0028509_session_1.stripped.N4.nii.gz
+ADNI1070047,M,61.4,adni_storage/ADNI_nii_gz_bias_corrected/I1070047_006_S_6610.stripped.N4.nii.gz
+ADNI10654619,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I10654619_100_S_10035.stripped.N4.nii.gz
+fcon1000Beijing_sub37602,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub37602.stripped.N4.nii.gz
+DLBS0028409_session_1,M,70.68,dlbs_storage/DLBS_bias_corrected/0028409_session_1.stripped.N4.nii.gz
+ADNI346866,F,56.9,adni_storage/ADNI_nii_gz_bias_corrected/I346866_128_S_4607.stripped.N4.nii.gz
+fcon1000AnnArbor_sub39923,M,70.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub39923.stripped.N4.nii.gz
+sub-CC420324,F,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420324.stripped.N4.nii.gz
+fcon1000Cleveland_sub92232,F,45.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub92232.stripped.N4.nii.gz
+fcon1000Baltimore_sub90658,F,32.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub90658.stripped.N4.nii.gz
+DLBS0028634_session_1,F,38.33,dlbs_storage/DLBS_bias_corrected/0028634_session_1.stripped.N4.nii.gz
+ADNI1536326,M,62.3,adni_storage/ADNI_nii_gz_bias_corrected/I1536326_052_S_7037.stripped.N4.nii.gz
+sub-CC510393,M,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510393.stripped.N4.nii.gz
+DLBS0028639_session_1,F,74.96,dlbs_storage/DLBS_bias_corrected/0028639_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub62843,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub62843.stripped.N4.nii.gz
+fcon1000Leiden_sub01553,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub01553.stripped.N4.nii.gz
+fcon1000Leiden_sub72247,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub72247.stripped.N4.nii.gz
+DLBS0028429_session_1,F,42.63,dlbs_storage/DLBS_bias_corrected/0028429_session_1.stripped.N4.nii.gz
+fcon1000Atlanta_sub72096,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub72096.stripped.N4.nii.gz
+fcon1000Cambridge_sub89894,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub89894.stripped.N4.nii.gz
+fcon1000Cambridge_sub69287,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub69287.stripped.N4.nii.gz
+fcon1000Beijing_sub88947,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub88947.stripped.N4.nii.gz
+sub-CC320429,M,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320429.stripped.N4.nii.gz
+fcon1000Cambridge_sub59434,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub59434.stripped.N4.nii.gz
+fcon1000Beijing_sub55541,M,25.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub55541.stripped.N4.nii.gz
+fcon1000Cambridge_sub42146,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub42146.stripped.N4.nii.gz
+sub-CC320107,M,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320107.stripped.N4.nii.gz
+ADNI10666922,F,64.6,adni_storage/ADNI_nii_gz_bias_corrected/I10666922_100_S_10091.stripped.N4.nii.gz
+DLBS0028588_session_1,M,48.53,dlbs_storage/DLBS_bias_corrected/0028588_session_1.stripped.N4.nii.gz
+sub-CC221324,M,36.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221324.stripped.N4.nii.gz
+DLBS0028361_session_1,M,55.56,dlbs_storage/DLBS_bias_corrected/0028361_session_1.stripped.N4.nii.gz
+ADNI10976428,F,64.2,adni_storage/ADNI_nii_gz_bias_corrected/I10976428_052_S_10275.stripped.N4.nii.gz
+sub-CC310214,F,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310214.stripped.N4.nii.gz
+DLBS0028360_session_1,F,54.03,dlbs_storage/DLBS_bias_corrected/0028360_session_1.stripped.N4.nii.gz
+sub-CC210519,F,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210519.stripped.N4.nii.gz
+fcon1000Cambridge_sub87846,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub87846.stripped.N4.nii.gz
+ADNI1514753,F,50.5,adni_storage/ADNI_nii_gz_bias_corrected/I1514753_037_S_7011.stripped.N4.nii.gz
+ADNI10889397,M,55.9,adni_storage/ADNI_nii_gz_bias_corrected/I10889397_019_S_10152.stripped.N4.nii.gz
+DLBS0028495_session_1,F,40.56,dlbs_storage/DLBS_bias_corrected/0028495_session_1.stripped.N4.nii.gz
+sub-CC620284,F,75.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620284.stripped.N4.nii.gz
+sub-CC210617,M,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210617.stripped.N4.nii.gz
+sub-CC420462,F,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420462.stripped.N4.nii.gz
+ADNI1151022,M,56.0,adni_storage/ADNI_nii_gz_bias_corrected/I1151022_016_S_6708.stripped.N4.nii.gz
+sub-CC420322,M,49.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420322.stripped.N4.nii.gz
+ADNI36696,F,56.4,adni_storage/ADNI_nii_gz_bias_corrected/I36696_016_S_1121.stripped.N4.nii.gz
+fcon1000AnnArbor_sub98007,F,66.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub98007.stripped.N4.nii.gz
+fcon1000Beijing_sub04191,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub04191.stripped.N4.nii.gz
+ADNI388757,M,64.3,adni_storage/ADNI_nii_gz_bias_corrected/I388757_116_S_4195.stripped.N4.nii.gz
+sub-CC610576,F,72.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610576.stripped.N4.nii.gz
+sub-CC720358,F,83.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720358.stripped.N4.nii.gz
+sub-CC420493,F,51.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420493.stripped.N4.nii.gz
+fcon1000Cambridge_sub81562,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub81562.stripped.N4.nii.gz
+DLBS0028408_session_1,F,20.7,dlbs_storage/DLBS_bias_corrected/0028408_session_1.stripped.N4.nii.gz
+fcon1000Leiden_sub39335,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub39335.stripped.N4.nii.gz
+ADNI907712,F,58.7,adni_storage/ADNI_nii_gz_bias_corrected/I907712_014_S_6076.stripped.N4.nii.gz
+fcon1000Beijing_sub56703,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub56703.stripped.N4.nii.gz
+ADNI313318,F,56.5,adni_storage/ADNI_nii_gz_bias_corrected/I313318_128_S_4607.stripped.N4.nii.gz
+sub-CC721888,F,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721888.stripped.N4.nii.gz
+ADNI324988,F,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I324988_023_S_4448.stripped.N4.nii.gz
+fcon1000Beijing_sub83624,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub83624.stripped.N4.nii.gz
+fcon1000Beijing_sub19974,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub19974.stripped.N4.nii.gz
+ADNI228879,F,56.0,adni_storage/ADNI_nii_gz_bias_corrected/I228879_031_S_4024.stripped.N4.nii.gz
+ADNI376810,F,59.4,adni_storage/ADNI_nii_gz_bias_corrected/I376810_067_S_5205.stripped.N4.nii.gz
+fcon1000Bangor_sub63767,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub63767.stripped.N4.nii.gz
+sub-CC320448,M,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320448.stripped.N4.nii.gz
+ADNI341795,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I341795_130_S_4982.stripped.N4.nii.gz
+ADNI140028,M,62.1,adni_storage/ADNI_nii_gz_bias_corrected/I140028_020_S_1288.stripped.N4.nii.gz
+sub-CC120795,M,24.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120795.stripped.N4.nii.gz
+ADNI382180,F,62.8,adni_storage/ADNI_nii_gz_bias_corrected/I382180_018_S_5240.stripped.N4.nii.gz
+fcon1000Beijing_sub69518,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub69518.stripped.N4.nii.gz
+sub-CC420720,F,50.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420720.stripped.N4.nii.gz
+ADNI1005735,M,55.3,adni_storage/ADNI_nii_gz_bias_corrected/I1005735_003_S_6264.stripped.N4.nii.gz
+fcon1000Cambridge_sub02953,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub02953.stripped.N4.nii.gz
+sub-CC610631,F,77.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610631.stripped.N4.nii.gz
+fcon1000Cambridge_sub17584,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub17584.stripped.N4.nii.gz
+sub-CC710591,F,85.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710591.stripped.N4.nii.gz
+fcon1000AnnArbor_sub33437,F,19.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub33437.stripped.N4.nii.gz
+sub-CC711035,F,88.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC711035.stripped.N4.nii.gz
+fcon1000Milwaukee_sub50771,F,53.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub50771.stripped.N4.nii.gz
+fcon1000Cleveland_sub64706,F,57.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub64706.stripped.N4.nii.gz
+ADNI1062510,F,62.6,adni_storage/ADNI_nii_gz_bias_corrected/I1062510_019_S_6585.stripped.N4.nii.gz
+sub-CC320218,F,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320218.stripped.N4.nii.gz
+sub-CC121144,M,26.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121144.stripped.N4.nii.gz
+sub-CC320269,F,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320269.stripped.N4.nii.gz
+fcon1000Beijing_sub50972,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub50972.stripped.N4.nii.gz
+DLBS0028593_session_1,F,62.06,dlbs_storage/DLBS_bias_corrected/0028593_session_1.stripped.N4.nii.gz
+ADNI1532649,F,59.9,adni_storage/ADNI_nii_gz_bias_corrected/I1532649_067_S_7029.stripped.N4.nii.gz
+fcon1000NewYork_sub08595,M,23.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub08595.stripped.N4.nii.gz
+fcon1000Newark_sub71042,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub71042.stripped.N4.nii.gz
+ADNI37256,M,62.2,adni_storage/ADNI_nii_gz_bias_corrected/I37256_067_S_1185.stripped.N4.nii.gz
+fcon1000Beijing_sub22890,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub22890.stripped.N4.nii.gz
+fcon1000Beijing_sub16091,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub16091.stripped.N4.nii.gz
+fcon1000NewYork_sub51309,F,27.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub51309.stripped.N4.nii.gz
+fcon1000NewYork_sub09539,F,21.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub09539.stripped.N4.nii.gz
+sub-CC321053,F,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321053.stripped.N4.nii.gz
+DLBS0028481_session_1,F,41.87,dlbs_storage/DLBS_bias_corrected/0028481_session_1.stripped.N4.nii.gz
+DLBS0028597_session_1,M,52.52,dlbs_storage/DLBS_bias_corrected/0028597_session_1.stripped.N4.nii.gz
+fcon1000Bangor_sub61908,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub61908.stripped.N4.nii.gz
+sub-CC410325,F,54.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410325.stripped.N4.nii.gz
+DLBS0028347_session_1,F,85.66,dlbs_storage/DLBS_bias_corrected/0028347_session_1.stripped.N4.nii.gz
+ADNI10245972,M,59.2,adni_storage/ADNI_nii_gz_bias_corrected/I10245972_033_S_6976.stripped.N4.nii.gz
+sub-CC221033,F,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221033.stripped.N4.nii.gz
+ADNI1221674,F,63.4,adni_storage/ADNI_nii_gz_bias_corrected/I1221674_013_S_6780.stripped.N4.nii.gz
+DLBS0028635_session_1,F,23.5,dlbs_storage/DLBS_bias_corrected/0028635_session_1.stripped.N4.nii.gz
+ADNI345563,F,63.0,adni_storage/ADNI_nii_gz_bias_corrected/I345563_019_S_5019.stripped.N4.nii.gz
+ADNI98972,M,63.9,adni_storage/ADNI_nii_gz_bias_corrected/I98972_016_S_0769.stripped.N4.nii.gz
+ADNI254248,M,62.3,adni_storage/ADNI_nii_gz_bias_corrected/I254248_116_S_4195.stripped.N4.nii.gz
+ADNI1154285,F,63.4,adni_storage/ADNI_nii_gz_bias_corrected/I1154285_011_S_6714.stripped.N4.nii.gz
+fcon1000Beijing_sub30616,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub30616.stripped.N4.nii.gz
+DLBS0028632_session_1,F,59.96,dlbs_storage/DLBS_bias_corrected/0028632_session_1.stripped.N4.nii.gz
+ADNI10461435,F,56.1,adni_storage/ADNI_nii_gz_bias_corrected/I10461435_022_S_10066.stripped.N4.nii.gz
+DLBS0028377_session_1,F,64.37,dlbs_storage/DLBS_bias_corrected/0028377_session_1.stripped.N4.nii.gz
+ADNI1433667,F,60.4,adni_storage/ADNI_nii_gz_bias_corrected/I1433667_067_S_6912.stripped.N4.nii.gz
+ADNI1669252,M,62.2,adni_storage/ADNI_nii_gz_bias_corrected/I1669252_035_S_7121.stripped.N4.nii.gz
+fcon1000Baltimore_sub52358,F,40.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub52358.stripped.N4.nii.gz
+sub-CC610071,F,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610071.stripped.N4.nii.gz
+sub-CC722536,F,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC722536.stripped.N4.nii.gz
+fcon1000NewYork_sub14299,F,24.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub14299.stripped.N4.nii.gz
+fcon1000Atlanta_sub81596,M,53.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub81596.stripped.N4.nii.gz
+fcon1000Beijing_sub28965,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub28965.stripped.N4.nii.gz
+sub-CC710486,F,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710486.stripped.N4.nii.gz
+fcon1000Cambridge_sub72068,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub72068.stripped.N4.nii.gz
+ADNI1143230,F,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I1143230_023_S_6702.stripped.N4.nii.gz
+DLBS0028573_session_1,M,44.29,dlbs_storage/DLBS_bias_corrected/0028573_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub43409,M,71.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub43409.stripped.N4.nii.gz
+fcon1000Cleveland_sub97844,F,30.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub97844.stripped.N4.nii.gz
+sub-CC620572,F,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620572.stripped.N4.nii.gz
+ADNI10954719,F,56.2,adni_storage/ADNI_nii_gz_bias_corrected/I10954719_052_S_10168.stripped.N4.nii.gz
+ADNI10989649,M,62.2,adni_storage/ADNI_nii_gz_bias_corrected/I10989649_019_S_10263.stripped.N4.nii.gz
+ADNI308620,M,57.0,adni_storage/ADNI_nii_gz_bias_corrected/I308620_014_S_4039.stripped.N4.nii.gz
+sub-CC721704,F,82.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721704.stripped.N4.nii.gz
+sub-CC510568,M,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510568.stripped.N4.nii.gz
+fcon1000Beijing_sub94536,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub94536.stripped.N4.nii.gz
+ADNI10392424,F,64.5,adni_storage/ADNI_nii_gz_bias_corrected/I10392424_035_S_7043.stripped.N4.nii.gz
+fcon1000Leiden_sub64642,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub64642.stripped.N4.nii.gz
+fcon1000AnnArbor_sub07921,F,19.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub07921.stripped.N4.nii.gz
+DLBS0028557_session_1,F,75.89,dlbs_storage/DLBS_bias_corrected/0028557_session_1.stripped.N4.nii.gz
+DLBS0028443_session_1,M,36.18,dlbs_storage/DLBS_bias_corrected/0028443_session_1.stripped.N4.nii.gz
+sub-CC721418,F,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721418.stripped.N4.nii.gz
+sub-CC221740,F,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221740.stripped.N4.nii.gz
+fcon1000Bangor_sub14388,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub14388.stripped.N4.nii.gz
+ADNI73491,F,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I73491_067_S_1253.stripped.N4.nii.gz
+sub-CC120218,F,27.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120218.stripped.N4.nii.gz
+fcon1000Beijing_sub28792,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub28792.stripped.N4.nii.gz
+fcon1000Beijing_sub17159,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub17159.stripped.N4.nii.gz
+ADNI1261605,F,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I1261605_016_S_6802.stripped.N4.nii.gz
+fcon1000Milwaukee_sub99479,F,65.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub99479.stripped.N4.nii.gz
+sub-CC320089,F,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320089.stripped.N4.nii.gz
+fcon1000AnnArbor_sub53959,M,20.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub53959.stripped.N4.nii.gz
+fcon1000Cambridge_sub25099,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub25099.stripped.N4.nii.gz
+fcon1000Cleveland_sub65858,F,53.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub65858.stripped.N4.nii.gz
+fcon1000Cambridge_sub10268,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub10268.stripped.N4.nii.gz
+fcon1000Beijing_sub22661,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub22661.stripped.N4.nii.gz
+sub-CC620413,M,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620413.stripped.N4.nii.gz
+ADNI1634180,F,58.0,adni_storage/ADNI_nii_gz_bias_corrected/I1634180_013_S_7103.stripped.N4.nii.gz
+ADNI44673,M,60.1,adni_storage/ADNI_nii_gz_bias_corrected/I44673_020_S_1288.stripped.N4.nii.gz
+fcon1000Beijing_sub48563,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub48563.stripped.N4.nii.gz
+sub-CC410387,M,57.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410387.stripped.N4.nii.gz
+fcon1000Beijing_sub89941,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub89941.stripped.N4.nii.gz
+DLBS0028630_session_1,M,38.37,dlbs_storage/DLBS_bias_corrected/0028630_session_1.stripped.N4.nii.gz
+sub-CC620821,M,70.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620821.stripped.N4.nii.gz
+ADNI10914005,F,59.1,adni_storage/ADNI_nii_gz_bias_corrected/I10914005_052_S_10185.stripped.N4.nii.gz
+ADNI1494601,F,58.7,adni_storage/ADNI_nii_gz_bias_corrected/I1494601_037_S_6995.stripped.N4.nii.gz
+fcon1000Cambridge_sub53296,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub53296.stripped.N4.nii.gz
+ADNI399834,F,62.3,adni_storage/ADNI_nii_gz_bias_corrected/I399834_137_S_0994.stripped.N4.nii.gz
+sub-CC420402,M,49.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420402.stripped.N4.nii.gz
+ADNI73381,F,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I73381_051_S_1072.stripped.N4.nii.gz
+fcon1000NewYork_sub10011,F,40.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub10011.stripped.N4.nii.gz
+fcon1000Atlanta_sub60499,M,29.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub60499.stripped.N4.nii.gz
+ADNI1584200,F,62.8,adni_storage/ADNI_nii_gz_bias_corrected/I1584200_035_S_7076.stripped.N4.nii.gz
+sub-CC720646,M,81.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720646.stripped.N4.nii.gz
+sub-CC221775,F,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221775.stripped.N4.nii.gz
+ADNI31766,F,64.6,adni_storage/ADNI_nii_gz_bias_corrected/I31766_023_S_1104.stripped.N4.nii.gz
+ADNI291512,M,62.9,adni_storage/ADNI_nii_gz_bias_corrected/I291512_116_S_4195.stripped.N4.nii.gz
+fcon1000Cambridge_sub11388,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub11388.stripped.N4.nii.gz
+DLBS0028437_session_1,F,64.01,dlbs_storage/DLBS_bias_corrected/0028437_session_1.stripped.N4.nii.gz
+DLBS0028533_session_1,F,46.77,dlbs_storage/DLBS_bias_corrected/0028533_session_1.stripped.N4.nii.gz
+ADNI974673,M,56.2,adni_storage/ADNI_nii_gz_bias_corrected/I974673_023_S_6270.stripped.N4.nii.gz
+fcon1000Cambridge_sub15258,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub15258.stripped.N4.nii.gz
+ADNI367181,M,61.8,adni_storage/ADNI_nii_gz_bias_corrected/I367181_135_S_4954.stripped.N4.nii.gz
+ADNI879215,M,56.0,adni_storage/ADNI_nii_gz_bias_corrected/I879215_114_S_6039.stripped.N4.nii.gz
+fcon1000NewYork_sub04856,M,9.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub04856.stripped.N4.nii.gz
+ADNI1604933,F,58.9,adni_storage/ADNI_nii_gz_bias_corrected/I1604933_067_S_7094.stripped.N4.nii.gz
+fcon1000Cambridge_sub62732,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub62732.stripped.N4.nii.gz
+sub-CC222652,M,30.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222652.stripped.N4.nii.gz
+sub-CC620405,F,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620405.stripped.N4.nii.gz
+sub-CC520197,F,59.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520197.stripped.N4.nii.gz
+fcon1000Cambridge_sub66351,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub66351.stripped.N4.nii.gz
+fcon1000Beijing_sub36942,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub36942.stripped.N4.nii.gz
+sub-CC520868,M,67.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520868.stripped.N4.nii.gz
+fcon1000AnnArbor_sub28433,M,21.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub28433.stripped.N4.nii.gz
+ADNI394453,M,62.3,adni_storage/ADNI_nii_gz_bias_corrected/I394453_135_S_4954.stripped.N4.nii.gz
+ADNI345558,F,63.0,adni_storage/ADNI_nii_gz_bias_corrected/I345558_019_S_5019.stripped.N4.nii.gz
+ADNI114057,F,57.9,adni_storage/ADNI_nii_gz_bias_corrected/I114057_016_S_1121.stripped.N4.nii.gz
+fcon1000Bangor_sub77520,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub77520.stripped.N4.nii.gz
+ADNI389098,F,59.6,adni_storage/ADNI_nii_gz_bias_corrected/I389098_067_S_5205.stripped.N4.nii.gz
+fcon1000Cambridge_sub27796,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub27796.stripped.N4.nii.gz
+fcon1000Cambridge_sub09633,F,30.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub09633.stripped.N4.nii.gz
+ADNI99328,M,61.1,adni_storage/ADNI_nii_gz_bias_corrected/I99328_020_S_1288.stripped.N4.nii.gz
+fcon1000Beijing_sub62438,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub62438.stripped.N4.nii.gz
+DLBS0028548_session_1,F,40.83,dlbs_storage/DLBS_bias_corrected/0028548_session_1.stripped.N4.nii.gz
+sub-CC420348,F,57.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420348.stripped.N4.nii.gz
+fcon1000NewYork_sub18638,M,23.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub18638.stripped.N4.nii.gz
+ADNI379689,M,60.9,adni_storage/ADNI_nii_gz_bias_corrected/I379689_131_S_5138.stripped.N4.nii.gz
+ADNI376818,F,59.4,adni_storage/ADNI_nii_gz_bias_corrected/I376818_067_S_5205.stripped.N4.nii.gz
+ADNI10443906,M,64.5,adni_storage/ADNI_nii_gz_bias_corrected/I10443906_389_S_10076.stripped.N4.nii.gz
+fcon1000Cambridge_sub41814,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub41814.stripped.N4.nii.gz
+sub-CC620451,M,75.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620451.stripped.N4.nii.gz
+ADNI705596,F,60.4,adni_storage/ADNI_nii_gz_bias_corrected/I705596_128_S_4607.stripped.N4.nii.gz
+sub-CC320850,F,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320850.stripped.N4.nii.gz
+sub-CC321428,F,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321428.stripped.N4.nii.gz
+sub-CC420071,F,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420071.stripped.N4.nii.gz
+ADNI335137,M,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I335137_073_S_4552.stripped.N4.nii.gz
+ADNI10932456,M,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I10932456_035_S_7121.stripped.N4.nii.gz
+fcon1000Milwaukee_sub73547,F,55.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub73547.stripped.N4.nii.gz
+ADNI1356513,F,64.6,adni_storage/ADNI_nii_gz_bias_corrected/I1356513_137_S_6883.stripped.N4.nii.gz
+ADNI1228298,F,61.9,adni_storage/ADNI_nii_gz_bias_corrected/I1228298_131_S_6805.stripped.N4.nii.gz
+fcon1000Cambridge_sub92393,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub92393.stripped.N4.nii.gz
+fcon1000Beijing_sub18960,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub18960.stripped.N4.nii.gz
+fcon1000NewYork_sub14465,M,47.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub14465.stripped.N4.nii.gz
+DLBS0028577_session_1,M,65.52,dlbs_storage/DLBS_bias_corrected/0028577_session_1.stripped.N4.nii.gz
+sub-CC310256,M,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310256.stripped.N4.nii.gz
+fcon1000Beijing_sub59347,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub59347.stripped.N4.nii.gz
+DLBS0028500_session_1,M,32.85,dlbs_storage/DLBS_bias_corrected/0028500_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub69696,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub69696.stripped.N4.nii.gz
+fcon1000Milwaukee_sub23607,F,48.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub23607.stripped.N4.nii.gz
+ADNI313538,M,64.1,adni_storage/ADNI_nii_gz_bias_corrected/I313538_116_S_4625.stripped.N4.nii.gz
+fcon1000NewYork_sub03951,M,37.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub03951.stripped.N4.nii.gz
+sub-CC310407,F,39.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310407.stripped.N4.nii.gz
+ADNI1260254,M,62.4,adni_storage/ADNI_nii_gz_bias_corrected/I1260254_006_S_6610.stripped.N4.nii.gz
+ADNI114262,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I114262_133_S_0913.stripped.N4.nii.gz
+ADNI365441,F,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I365441_073_S_5023.stripped.N4.nii.gz
+DLBS0028628_session_1,M,51.81,dlbs_storage/DLBS_bias_corrected/0028628_session_1.stripped.N4.nii.gz
+ADNI705606,F,60.4,adni_storage/ADNI_nii_gz_bias_corrected/I705606_128_S_4607.stripped.N4.nii.gz
+fcon1000Bangor_sub61418,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub61418.stripped.N4.nii.gz
+DLBS0028623_session_1,F,30.31,dlbs_storage/DLBS_bias_corrected/0028623_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub47278,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub47278.stripped.N4.nii.gz
+sub-CC110056,F,22.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110056.stripped.N4.nii.gz
+fcon1000Cambridge_sub47231,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub47231.stripped.N4.nii.gz
+sub-CC510648,F,61.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510648.stripped.N4.nii.gz
+sub-CC210051,F,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210051.stripped.N4.nii.gz
+fcon1000Berlin_sub40143,M,34.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub40143.stripped.N4.nii.gz
+ADNI367565,M,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I367565_131_S_5138.stripped.N4.nii.gz
+DLBS0028439_session_1,M,25.38,dlbs_storage/DLBS_bias_corrected/0028439_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub55736,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub55736.stripped.N4.nii.gz
+ADNI10908346,M,60.1,adni_storage/ADNI_nii_gz_bias_corrected/I10908346_073_S_6925.stripped.N4.nii.gz
+DLBS0028615_session_1,F,80.95,dlbs_storage/DLBS_bias_corrected/0028615_session_1.stripped.N4.nii.gz
+DLBS0028326_session_1,M,66.42,dlbs_storage/DLBS_bias_corrected/0028326_session_1.stripped.N4.nii.gz
+ADNI31765,F,64.6,adni_storage/ADNI_nii_gz_bias_corrected/I31765_023_S_1104.stripped.N4.nii.gz
+DLBS0028553_session_1,M,30.76,dlbs_storage/DLBS_bias_corrected/0028553_session_1.stripped.N4.nii.gz
+ADNI1423364,F,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I1423364_003_S_6915.stripped.N4.nii.gz
+DLBS0028470_session_1,F,43.19,dlbs_storage/DLBS_bias_corrected/0028470_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub73168,M,78.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub73168.stripped.N4.nii.gz
+sub-CC310331,M,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310331.stripped.N4.nii.gz
+ADNI10952478,F,55.6,adni_storage/ADNI_nii_gz_bias_corrected/I10952478_022_S_10186.stripped.N4.nii.gz
+ADNI1032530,F,55.9,adni_storage/ADNI_nii_gz_bias_corrected/I1032530_116_S_6550.stripped.N4.nii.gz
+ADNI10939739,F,62.5,adni_storage/ADNI_nii_gz_bias_corrected/I10939739_301_S_10230.stripped.N4.nii.gz
+fcon1000Cambridge_sub88445,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub88445.stripped.N4.nii.gz
+DLBS0028582_session_1,M,48.45,dlbs_storage/DLBS_bias_corrected/0028582_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub09397,M,30.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub09397.stripped.N4.nii.gz
+sub-CC222264,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222264.stripped.N4.nii.gz
+fcon1000Leiden_sub19281,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub19281.stripped.N4.nii.gz
+ADNI1017357,F,59.0,adni_storage/ADNI_nii_gz_bias_corrected/I1017357_024_S_6033.stripped.N4.nii.gz
+fcon1000Baltimore_sub86414,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub86414.stripped.N4.nii.gz
+fcon1000Beijing_sub72678,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub72678.stripped.N4.nii.gz
+fcon1000Berlin_sub06204,M,34.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub06204.stripped.N4.nii.gz
+ADNI335031,M,63.4,adni_storage/ADNI_nii_gz_bias_corrected/I335031_116_S_4195.stripped.N4.nii.gz
+sub-CC510548,M,62.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510548.stripped.N4.nii.gz
+sub-CC220394,F,33.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220394.stripped.N4.nii.gz
+DLBS0028523_session_1,M,49.68,dlbs_storage/DLBS_bias_corrected/0028523_session_1.stripped.N4.nii.gz
+fcon1000NewYork_sub35262,F,28.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub35262.stripped.N4.nii.gz
+sub-CC520584,F,59.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520584.stripped.N4.nii.gz
+fcon1000Beijing_sub43290,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub43290.stripped.N4.nii.gz
+fcon1000Cambridge_sub39053,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub39053.stripped.N4.nii.gz
+DLBS0028388_session_1,M,82.93,dlbs_storage/DLBS_bias_corrected/0028388_session_1.stripped.N4.nii.gz
+sub-CC320651,M,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320651.stripped.N4.nii.gz
+fcon1000Beijing_sub95575,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub95575.stripped.N4.nii.gz
+fcon1000Cambridge_sub84256,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub84256.stripped.N4.nii.gz
+sub-CC520377,F,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520377.stripped.N4.nii.gz
+fcon1000NewYork_sub30623,F,26.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub30623.stripped.N4.nii.gz
+fcon1000Beijing_sub80551,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub80551.stripped.N4.nii.gz
+sub-CC510043,M,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510043.stripped.N4.nii.gz
+sub-CC720238,F,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720238.stripped.N4.nii.gz
+fcon1000Baltimore_sub85922,M,30.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub85922.stripped.N4.nii.gz
+fcon1000Munchen_sub50162,M,73.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub50162.stripped.N4.nii.gz
+DLBS0028518_session_1,F,71.09,dlbs_storage/DLBS_bias_corrected/0028518_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub24757,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub24757.stripped.N4.nii.gz
+fcon1000Beijing_sub46259,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub46259.stripped.N4.nii.gz
+sub-CC420589,M,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420589.stripped.N4.nii.gz
+ADNI37607,F,60.1,adni_storage/ADNI_nii_gz_bias_corrected/I37607_051_S_1072.stripped.N4.nii.gz
+fcon1000NewYork_sub44979,F,15.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub44979.stripped.N4.nii.gz
+sub-CC620314,F,74.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620314.stripped.N4.nii.gz
+fcon1000Cambridge_sub55660,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub55660.stripped.N4.nii.gz
+sub-CC320359,M,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320359.stripped.N4.nii.gz
+sub-CC320608,M,45.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320608.stripped.N4.nii.gz
+DLBS0028595_session_1,F,87.68,dlbs_storage/DLBS_bias_corrected/0028595_session_1.stripped.N4.nii.gz
+sub-CC620193,M,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620193.stripped.N4.nii.gz
+fcon1000Cambridge_sub51671,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub51671.stripped.N4.nii.gz
+sub-CC721374,M,86.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721374.stripped.N4.nii.gz
+fcon1000Cambridge_sub81524,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub81524.stripped.N4.nii.gz
+sub-CC221040,M,36.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221040.stripped.N4.nii.gz
+sub-CC310361,F,43.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310361.stripped.N4.nii.gz
+fcon1000Cambridge_sub45344,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub45344.stripped.N4.nii.gz
+ADNI1484664,F,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I1484664_067_S_6525.stripped.N4.nii.gz
+DLBS0028334_session_1,F,55.26,dlbs_storage/DLBS_bias_corrected/0028334_session_1.stripped.N4.nii.gz
+ADNI10951691,F,59.0,adni_storage/ADNI_nii_gz_bias_corrected/I10951691_014_S_10236.stripped.N4.nii.gz
+fcon1000Munchen_sub74607,M,65.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub74607.stripped.N4.nii.gz
+fcon1000Milwaukee_sub16666,F,47.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub16666.stripped.N4.nii.gz
+ADNI1501367,F,56.6,adni_storage/ADNI_nii_gz_bias_corrected/I1501367_035_S_7004.stripped.N4.nii.gz
+fcon1000Cleveland_sub28596,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub28596.stripped.N4.nii.gz
+DLBS0028466_session_1,M,74.5,dlbs_storage/DLBS_bias_corrected/0028466_session_1.stripped.N4.nii.gz
+ADNI60862,F,57.3,adni_storage/ADNI_nii_gz_bias_corrected/I60862_133_S_0913.stripped.N4.nii.gz
+sub-CC510284,F,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510284.stripped.N4.nii.gz
+sub-CC620526,F,73.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620526.stripped.N4.nii.gz
+fcon1000Cambridge_sub64985,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub64985.stripped.N4.nii.gz
+fcon1000NewYork_sub37864,F,41.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub37864.stripped.N4.nii.gz
+sub-CC620935,M,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620935.stripped.N4.nii.gz
+sub-CC610285,M,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610285.stripped.N4.nii.gz
+ADNI291506,M,62.9,adni_storage/ADNI_nii_gz_bias_corrected/I291506_116_S_4195.stripped.N4.nii.gz
+sub-CC320109,M,39.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320109.stripped.N4.nii.gz
+sub-CC620359,M,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620359.stripped.N4.nii.gz
+ADNI1246020,F,61.5,adni_storage/ADNI_nii_gz_bias_corrected/I1246020_003_S_6479.stripped.N4.nii.gz
+sub-CC420392,F,51.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420392.stripped.N4.nii.gz
+sub-CC222304,M,38.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222304.stripped.N4.nii.gz
+ADNI339295,M,63.5,adni_storage/ADNI_nii_gz_bias_corrected/I339295_073_S_4552.stripped.N4.nii.gz
+fcon1000Atlanta_sub91049,F,48.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub91049.stripped.N4.nii.gz
+DLBS0028637_session_1,F,44.54,dlbs_storage/DLBS_bias_corrected/0028637_session_1.stripped.N4.nii.gz
+ADNI1267895,M,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I1267895_041_S_6136.stripped.N4.nii.gz
+DLBS0028363_session_1,M,26.02,dlbs_storage/DLBS_bias_corrected/0028363_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub39065,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub39065.stripped.N4.nii.gz
+sub-CC510486,F,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510486.stripped.N4.nii.gz
+fcon1000Cambridge_sub08588,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub08588.stripped.N4.nii.gz
+ADNI1092176,F,56.6,adni_storage/ADNI_nii_gz_bias_corrected/I1092176_019_S_6635.stripped.N4.nii.gz
+ADNI256318,M,56.3,adni_storage/ADNI_nii_gz_bias_corrected/I256318_014_S_4039.stripped.N4.nii.gz
+sub-CC223085,M,38.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC223085.stripped.N4.nii.gz
+fcon1000Atlanta_sub06870,M,57.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub06870.stripped.N4.nii.gz
+sub-CC720670,F,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720670.stripped.N4.nii.gz
+DLBS0028392_session_1,F,54.85,dlbs_storage/DLBS_bias_corrected/0028392_session_1.stripped.N4.nii.gz
+fcon1000Berlin_sub28092,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub28092.stripped.N4.nii.gz
+fcon1000AnnArbor_sub82334,M,14.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub82334.stripped.N4.nii.gz
+fcon1000Beijing_sub98617,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub98617.stripped.N4.nii.gz
+fcon1000Cambridge_sub58874,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub58874.stripped.N4.nii.gz
+fcon1000Beijing_sub68597,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub68597.stripped.N4.nii.gz
+ADNI90755,F,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I90755_051_S_1072.stripped.N4.nii.gz
+sub-CC620264,F,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620264.stripped.N4.nii.gz
+DLBS0028400_session_1,M,49.92,dlbs_storage/DLBS_bias_corrected/0028400_session_1.stripped.N4.nii.gz
+DLBS0028607_session_1,M,66.74,dlbs_storage/DLBS_bias_corrected/0028607_session_1.stripped.N4.nii.gz
+fcon1000Leiden_sub86034,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub86034.stripped.N4.nii.gz
+fcon1000Beijing_sub74587,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub74587.stripped.N4.nii.gz
+DLBS0028513_session_1,M,26.68,dlbs_storage/DLBS_bias_corrected/0028513_session_1.stripped.N4.nii.gz
+sub-CC320661,F,43.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320661.stripped.N4.nii.gz
+sub-CC121479,F,26.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121479.stripped.N4.nii.gz
+ADNI1236085,F,61.7,adni_storage/ADNI_nii_gz_bias_corrected/I1236085_022_S_6822.stripped.N4.nii.gz
+sub-CC321073,F,43.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321073.stripped.N4.nii.gz
+DLBS0028578_session_1,M,87.34,dlbs_storage/DLBS_bias_corrected/0028578_session_1.stripped.N4.nii.gz
+ADNI871835,F,57.7,adni_storage/ADNI_nii_gz_bias_corrected/I871835_141_S_6041.stripped.N4.nii.gz
+sub-CC722651,F,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC722651.stripped.N4.nii.gz
+fcon1000Cambridge_sub52036,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub52036.stripped.N4.nii.gz
+ADNI228854,F,63.5,adni_storage/ADNI_nii_gz_bias_corrected/I228854_037_S_4028.stripped.N4.nii.gz
+ADNI916399,F,55.8,adni_storage/ADNI_nii_gz_bias_corrected/I916399_168_S_6085.stripped.N4.nii.gz
+ADNI982860,F,62.6,adni_storage/ADNI_nii_gz_bias_corrected/I982860_109_S_6218.stripped.N4.nii.gz
+fcon1000NewYork_sub77903,M,24.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub77903.stripped.N4.nii.gz
+ADNI367563,M,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I367563_131_S_5138.stripped.N4.nii.gz
+fcon1000Beijing_sub65467,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub65467.stripped.N4.nii.gz
+sub-CC520011,F,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520011.stripped.N4.nii.gz
+fcon1000Beijing_sub72654,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub72654.stripped.N4.nii.gz
+sub-CC120008,M,26.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120008.stripped.N4.nii.gz
+fcon1000Cambridge_sub27065,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub27065.stripped.N4.nii.gz
+fcon1000Beijing_sub01244,F,25.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub01244.stripped.N4.nii.gz
+sub-CC711128,M,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC711128.stripped.N4.nii.gz
+sub-CC110606,M,20.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110606.stripped.N4.nii.gz
+fcon1000AnnArbor_sub97518,F,21.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub97518.stripped.N4.nii.gz
+fcon1000Atlanta_sub71337,M,27.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub71337.stripped.N4.nii.gz
diff --git a/combined_metadata.csv b/combined_metadata.csv
new file mode 100644
index 0000000..63a8fc7
--- /dev/null
+++ b/combined_metadata.csv
@@ -0,0 +1,1750 @@
+indx,Project,uid,age_at_scan,partition,path
+0,IXI,IXI002,35.8,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI002-Guys-0828-T1.nii.gz
+1,IXI,IXI012,38.78,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI012-HH-1211-T1.nii.gz
+2,IXI,IXI013,46.71,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI013-HH-1212-T1.nii.gz
+3,IXI,IXI014,34.24,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI014-HH-1236-T1.nii.gz
+4,IXI,IXI015,24.28,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI015-HH-1258-T1.nii.gz
+5,IXI,IXI016,55.17,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI016-Guys-0697-T1.nii.gz
+6,IXI,IXI017,29.09,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI017-Guys-0698-T1.nii.gz
+7,IXI,IXI019,58.66,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI019-Guys-0702-T1.nii.gz
+8,IXI,IXI020,39.47,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI020-Guys-0700-T1.nii.gz
+9,IXI,IXI021,21.57,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI021-Guys-0703-T1.nii.gz
+10,IXI,IXI022,30.67,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI022-Guys-0701-T1.nii.gz
+11,IXI,IXI023,33.08,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI023-Guys-0699-T1.nii.gz
+12,IXI,IXI024,37.7,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI024-Guys-0705-T1.nii.gz
+13,IXI,IXI025,29.78,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI025-Guys-0852-T1.nii.gz
+14,IXI,IXI026,34.48,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI026-Guys-0696-T1.nii.gz
+15,IXI,IXI027,30.42,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI027-Guys-0710-T1.nii.gz
+16,IXI,IXI028,74.03,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI028-Guys-1038-T1.nii.gz
+17,IXI,IXI029,59.22,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI029-Guys-0829-T1.nii.gz
+18,IXI,IXI030,31.46,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI030-Guys-0708-T1.nii.gz
+19,IXI,IXI031,31.13,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI031-Guys-0797-T1.nii.gz
+20,IXI,IXI033,24.76,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI033-HH-1259-T1.nii.gz
+21,IXI,IXI034,24.34,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI034-HH-1260-T1.nii.gz
+22,IXI,IXI035,37.14,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI035-IOP-0873-T1.nii.gz
+23,IXI,IXI036,23.54,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI036-Guys-0736-T1.nii.gz
+24,IXI,IXI037,37.21,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI037-Guys-0704-T1.nii.gz
+25,IXI,IXI038,21.38,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI038-Guys-0729-T1.nii.gz
+26,IXI,IXI039,23.35,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI039-HH-1261-T1.nii.gz
+27,IXI,IXI040,44.09,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI040-Guys-0724-T1.nii.gz
+28,IXI,IXI041,27.39,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI041-Guys-0706-T1.nii.gz
+29,IXI,IXI042,27.82,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI042-Guys-0725-T1.nii.gz
+30,IXI,IXI043,22.65,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI043-Guys-0714-T1.nii.gz
+31,IXI,IXI044,44.85,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI044-Guys-0712-T1.nii.gz
+32,IXI,IXI045,40.04,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI045-Guys-0713-T1.nii.gz
+33,IXI,IXI046,37.73,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI046-Guys-0824-T1.nii.gz
+34,IXI,IXI048,50.65,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI048-HH-1326-T1.nii.gz
+35,IXI,IXI049,31.85,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI049-HH-1358-T1.nii.gz
+36,IXI,IXI050,63.19,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI050-Guys-0711-T1.nii.gz
+37,IXI,IXI051,25.74,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI051-HH-1328-T1.nii.gz
+38,IXI,IXI052,33.76,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI052-HH-1343-T1.nii.gz
+39,IXI,IXI053,53.31,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI053-Guys-0727-T1.nii.gz
+40,IXI,IXI054,41.75,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI054-Guys-0707-T1.nii.gz
+41,IXI,IXI055,49.4,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI055-Guys-0730-T1.nii.gz
+42,IXI,IXI056,25.94,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI056-HH-1327-T1.nii.gz
+43,IXI,IXI057,28.16,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI057-HH-1342-T1.nii.gz
+44,IXI,IXI058,29.28,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI058-Guys-0726-T1.nii.gz
+45,IXI,IXI059,34.14,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI059-HH-1284-T1.nii.gz
+46,IXI,IXI060,30.7,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI060-Guys-0709-T1.nii.gz
+47,IXI,IXI061,59.0,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI061-Guys-0715-T1.nii.gz
+48,IXI,IXI062,36.22,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI062-Guys-0740-T1.nii.gz
+49,IXI,IXI063,41.12,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI063-Guys-0742-T1.nii.gz
+50,IXI,IXI064,20.07,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI064-Guys-0743-T1.nii.gz
+51,IXI,IXI065,23.72,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI065-Guys-0744-T1.nii.gz
+52,IXI,IXI066,46.17,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI066-Guys-0731-T1.nii.gz
+53,IXI,IXI067,24.69,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI067-HH-1356-T1.nii.gz
+54,IXI,IXI068,24.57,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI068-Guys-0756-T1.nii.gz
+55,IXI,IXI069,26.27,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI069-Guys-0769-T1.nii.gz
+56,IXI,IXI070,20.7,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI070-Guys-0767-T1.nii.gz
+57,IXI,IXI071,37.0,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI071-Guys-0770-T1.nii.gz
+58,IXI,IXI072,68.6,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI072-HH-2324-T1.nii.gz
+59,IXI,IXI073,46.37,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI073-Guys-0755-T1.nii.gz
+60,IXI,IXI074,22.53,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI074-Guys-0771-T1.nii.gz
+61,IXI,IXI075,35.99,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI075-Guys-0754-T1.nii.gz
+62,IXI,IXI076,55.83,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI076-Guys-0753-T1.nii.gz
+63,IXI,IXI077,36.48,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI077-Guys-0752-T1.nii.gz
+64,IXI,IXI078,28.51,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI078-Guys-0751-T1.nii.gz
+65,IXI,IXI079,52.53,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI079-HH-1388-T1.nii.gz
+66,IXI,IXI080,21.2,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI080-HH-1341-T1.nii.gz
+67,IXI,IXI083,30.89,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI083-HH-1357-T1.nii.gz
+68,IXI,IXI084,38.23,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI084-Guys-0741-T1.nii.gz
+69,IXI,IXI085,31.85,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI085-Guys-0759-T1.nii.gz
+70,IXI,IXI086,30.07,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI086-Guys-0728-T1.nii.gz
+71,IXI,IXI087,25.66,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI087-Guys-0768-T1.nii.gz
+72,IXI,IXI089,49.4,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI089-Guys-0757-T1.nii.gz
+73,IXI,IXI090,41.82,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI090-Guys-0800-T1.nii.gz
+74,IXI,IXI091,28.61,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI091-Guys-0762-T1.nii.gz
+75,IXI,IXI092,33.24,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI092-HH-1436-T1.nii.gz
+76,IXI,IXI093,26.88,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI093-HH-1359-T1.nii.gz
+77,IXI,IXI094,24.85,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI094-HH-1355-T1.nii.gz
+78,IXI,IXI095,24.9,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI095-HH-1390-T1.nii.gz
+79,IXI,IXI096,26.92,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI096-HH-1391-T1.nii.gz
+80,IXI,IXI097,35.02,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI097-HH-1619-T1.nii.gz
+81,IXI,IXI098,29.73,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI098-Guys-0745-T1.nii.gz
+82,IXI,IXI099,51.75,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI099-Guys-0748-T1.nii.gz
+83,IXI,IXI100,41.51,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI100-Guys-0747-T1.nii.gz
+84,IXI,IXI101,45.77,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI101-Guys-0749-T1.nii.gz
+85,IXI,IXI102,24.62,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI102-HH-1416-T1.nii.gz
+86,IXI,IXI103,31.81,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI103-Guys-0750-T1.nii.gz
+87,IXI,IXI104,26.98,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI104-HH-1450-T1.nii.gz
+88,IXI,IXI105,38.9,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI105-HH-1471-T1.nii.gz
+89,IXI,IXI106,36.67,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI106-Guys-0760-T1.nii.gz
+90,IXI,IXI107,31.9,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI107-Guys-0761-T1.nii.gz
+91,IXI,IXI108,43.16,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI108-Guys-0865-T1.nii.gz
+92,IXI,IXI109,36.47,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI109-Guys-0732-T1.nii.gz
+93,IXI,IXI110,37.77,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI110-Guys-0733-T1.nii.gz
+94,IXI,IXI111,25.19,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI111-Guys-0734-T1.nii.gz
+95,IXI,IXI112,23.7,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI112-Guys-0735-T1.nii.gz
+96,IXI,IXI113,32.36,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI113-Guys-0776-T1.nii.gz
+97,IXI,IXI114,65.41,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI114-Guys-0737-T1.nii.gz
+98,IXI,IXI115,72.13,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI115-Guys-0738-T1.nii.gz
+99,IXI,IXI116,23.85,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI116-Guys-0739-T1.nii.gz
+100,IXI,IXI118,25.93,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI118-Guys-0764-T1.nii.gz
+101,IXI,IXI119,32.17,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI119-Guys-0765-T1.nii.gz
+102,IXI,IXI120,53.67,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI120-Guys-0766-T1.nii.gz
+103,IXI,IXI121,35.07,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI121-Guys-0772-T1.nii.gz
+104,IXI,IXI122,22.97,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI122-Guys-0773-T1.nii.gz
+105,IXI,IXI123,46.58,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI123-Guys-0774-T1.nii.gz
+106,IXI,IXI126,57.22,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI126-HH-1437-T1.nii.gz
+107,IXI,IXI127,29.5,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI127-HH-1451-T1.nii.gz
+108,IXI,IXI128,27.33,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI128-HH-1470-T1.nii.gz
+109,IXI,IXI129,23.04,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI129-Guys-0775-T1.nii.gz
+110,IXI,IXI130,38.77,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI130-HH-1528-T1.nii.gz
+111,IXI,IXI131,45.23,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI131-HH-1527-T1.nii.gz
+112,IXI,IXI132,58.72,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI132-HH-1415-T1.nii.gz
+113,IXI,IXI134,47.35,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI134-Guys-0780-T1.nii.gz
+114,IXI,IXI135,29.0,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI135-Guys-0779-T1.nii.gz
+115,IXI,IXI136,35.81,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI136-HH-1452-T1.nii.gz
+116,IXI,IXI137,41.23,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI137-HH-1472-T1.nii.gz
+117,IXI,IXI138,33.14,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI138-Guys-0746-T1.nii.gz
+118,IXI,IXI139,32.94,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI139-Guys-0815-T1.nii.gz
+119,IXI,IXI140,27.84,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI140-Guys-0787-T1.nii.gz
+120,IXI,IXI141,46.01,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI141-Guys-0789-T1.nii.gz
+121,IXI,IXI142,48.25,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI142-Guys-0786-T1.nii.gz
+122,IXI,IXI143,31.96,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI143-Guys-0785-T1.nii.gz
+123,IXI,IXI144,29.36,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI144-Guys-0788-T1.nii.gz
+124,IXI,IXI145,49.39,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI145-Guys-0781-T1.nii.gz
+125,IXI,IXI146,33.59,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI146-HH-1389-T1.nii.gz
+126,IXI,IXI148,40.56,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI148-HH-1453-T1.nii.gz
+127,IXI,IXI150,27.01,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI150-HH-1550-T1.nii.gz
+128,IXI,IXI151,45.39,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI151-Guys-0793-T1.nii.gz
+129,IXI,IXI153,22.43,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI153-Guys-0782-T1.nii.gz
+130,IXI,IXI154,24.34,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI154-Guys-0821-T1.nii.gz
+131,IXI,IXI156,59.38,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI156-Guys-0837-T1.nii.gz
+132,IXI,IXI157,54.43,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI157-Guys-0816-T1.nii.gz
+133,IXI,IXI158,75.63,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI158-Guys-0783-T1.nii.gz
+134,IXI,IXI159,24.14,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI159-HH-1549-T1.nii.gz
+135,IXI,IXI160,39.4,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI160-HH-1637-T1.nii.gz
+136,IXI,IXI161,65.11,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI161-HH-2533-T1.nii.gz
+137,IXI,IXI162,50.52,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI162-HH-1548-T1.nii.gz
+138,IXI,IXI163,48.05,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI163-HH-1621-T1.nii.gz
+139,IXI,IXI164,64.47,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI164-Guys-0844-T1.nii.gz
+140,IXI,IXI165,24.38,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI165-HH-1589-T1.nii.gz
+141,IXI,IXI166,41.2,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI166-Guys-0846-T1.nii.gz
+142,IXI,IXI167,67.53,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI167-HH-1569-T1.nii.gz
+143,IXI,IXI168,43.82,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI168-HH-1607-T1.nii.gz
+144,IXI,IXI169,58.77,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI169-Guys-0842-T1.nii.gz
+145,IXI,IXI170,31.59,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI170-Guys-0843-T1.nii.gz
+146,IXI,IXI172,74.99,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI172-Guys-0982-T1.nii.gz
+147,IXI,IXI173,29.57,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI173-HH-1590-T1.nii.gz
+148,IXI,IXI174,63.1,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI174-HH-1571-T1.nii.gz
+149,IXI,IXI175,72.27,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI175-HH-1570-T1.nii.gz
+150,IXI,IXI176,24.85,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI176-HH-1604-T1.nii.gz
+151,IXI,IXI177,34.29,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI177-Guys-0831-T1.nii.gz
+152,IXI,IXI178,25.41,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI178-Guys-0778-T1.nii.gz
+153,IXI,IXI179,30.14,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI179-Guys-0777-T1.nii.gz
+154,IXI,IXI180,42.15,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI180-HH-1605-T1.nii.gz
+155,IXI,IXI181,45.7,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI181-Guys-0790-T1.nii.gz
+156,IXI,IXI182,60.97,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI182-Guys-0792-T1.nii.gz
+157,IXI,IXI183,30.18,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI183-Guys-0791-T1.nii.gz
+158,IXI,IXI184,30.12,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI184-Guys-0794-T1.nii.gz
+159,IXI,IXI185,83.46,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI185-Guys-0795-T1.nii.gz
+160,IXI,IXI186,61.15,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI186-Guys-0796-T1.nii.gz
+161,IXI,IXI188,44.02,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI188-Guys-0798-T1.nii.gz
+162,IXI,IXI189,41.72,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI189-Guys-0799-T1.nii.gz
+163,IXI,IXI191,44.21,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI191-Guys-0801-T1.nii.gz
+164,IXI,IXI192,38.94,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI192-Guys-0878-T1.nii.gz
+165,IXI,IXI193,55.24,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI193-Guys-0810-T1.nii.gz
+166,IXI,IXI194,48.33,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI194-Guys-0818-T1.nii.gz
+167,IXI,IXI195,37.83,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI195-HH-1620-T1.nii.gz
+168,IXI,IXI196,29.11,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI196-Guys-0805-T1.nii.gz
+169,IXI,IXI197,60.46,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI197-Guys-0811-T1.nii.gz
+170,IXI,IXI198,45.26,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI198-Guys-0803-T1.nii.gz
+171,IXI,IXI199,63.66,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI199-Guys-0802-T1.nii.gz
+172,IXI,IXI200,49.92,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI200-Guys-0812-T1.nii.gz
+173,IXI,IXI201,22.41,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI201-HH-1588-T1.nii.gz
+174,IXI,IXI202,20.17,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI202-HH-1526-T1.nii.gz
+175,IXI,IXI204,74.64,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI204-HH-1651-T1.nii.gz
+176,IXI,IXI205,45.38,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI205-HH-1649-T1.nii.gz
+177,IXI,IXI206,50.02,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI206-HH-1650-T1.nii.gz
+178,IXI,IXI207,40.36,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI207-Guys-0809-T1.nii.gz
+179,IXI,IXI208,34.0,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI208-Guys-0808-T1.nii.gz
+180,IXI,IXI209,32.64,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI209-Guys-0804-T1.nii.gz
+181,IXI,IXI210,69.33,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI210-Guys-0856-T1.nii.gz
+182,IXI,IXI211,55.12,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI211-HH-1568-T1.nii.gz
+183,IXI,IXI212,59.13,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI212-HH-1643-T1.nii.gz
+184,IXI,IXI213,54.66,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI213-HH-1642-T1.nii.gz
+185,IXI,IXI214,24.31,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI214-HH-1636-T1.nii.gz
+186,IXI,IXI216,73.02,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI216-HH-1635-T1.nii.gz
+187,IXI,IXI217,57.21,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI217-HH-1638-T1.nii.gz
+188,IXI,IXI218,35.41,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI218-HH-1815-T1.nii.gz
+189,IXI,IXI219,53.08,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI219-Guys-0894-T1.nii.gz
+190,IXI,IXI219,58.07,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI219-Guys-0894-T1.nii.gz
+191,IXI,IXI221,48.55,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI221-HH-1606-T1.nii.gz
+192,IXI,IXI222,36.06,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI222-Guys-0819-T1.nii.gz
+193,IXI,IXI223,63.25,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI223-Guys-0830-T1.nii.gz
+194,IXI,IXI224,36.92,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI224-Guys-0823-T1.nii.gz
+195,IXI,IXI225,47.01,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI225-Guys-0832-T1.nii.gz
+196,IXI,IXI226,41.9,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI226-HH-1618-T1.nii.gz
+197,IXI,IXI227,67.06,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI227-Guys-0813-T1.nii.gz
+198,IXI,IXI229,72.35,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI229-Guys-0980-T1.nii.gz
+199,IXI,IXI230,21.15,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI230-IOP-0869-T1.nii.gz
+200,IXI,IXI231,58.99,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI231-IOP-0866-T1.nii.gz
+201,IXI,IXI232,28.81,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI232-IOP-0898-T1.nii.gz
+202,IXI,IXI233,26.75,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI233-IOP-0875-T1.nii.gz
+203,IXI,IXI234,33.13,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI234-IOP-0870-T1.nii.gz
+204,IXI,IXI236,25.71,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI236-Guys-0817-T1.nii.gz
+205,IXI,IXI237,78.2,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI237-Guys-1049-T1.nii.gz
+206,IXI,IXI238,27.73,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI238-IOP-0883-T1.nii.gz
+207,IXI,IXI239,62.96,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI239-HH-2296-T1.nii.gz
+208,IXI,IXI240,39.04,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI240-Guys-0834-T1.nii.gz
+209,IXI,IXI241,30.01,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI241-Guys-0833-T1.nii.gz
+210,IXI,IXI242,31.37,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI242-HH-1722-T1.nii.gz
+211,IXI,IXI244,49.04,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI244-Guys-0841-T1.nii.gz
+212,IXI,IXI246,42.22,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI246-Guys-0840-T1.nii.gz
+213,IXI,IXI247,29.82,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI247-Guys-0838-T1.nii.gz
+214,IXI,IXI248,44.88,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI248-HH-1972-T1.nii.gz
+215,IXI,IXI249,68.27,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI249-Guys-1072-T1.nii.gz
+216,IXI,IXI250,72.88,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI250-Guys-0836-T1.nii.gz
+217,IXI,IXI251,80.17,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI251-Guys-1055-T1.nii.gz
+218,IXI,IXI252,70.56,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI252-HH-1693-T1.nii.gz
+219,IXI,IXI253,24.0,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI253-HH-1694-T1.nii.gz
+220,IXI,IXI254,25.58,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI254-HH-1705-T1.nii.gz
+221,IXI,IXI255,57.7,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI255-HH-1882-T1.nii.gz
+222,IXI,IXI256,27.38,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI256-HH-1723-T1.nii.gz
+223,IXI,IXI257,69.55,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI257-HH-1724-T1.nii.gz
+224,IXI,IXI258,22.57,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI258-HH-1769-T1.nii.gz
+225,IXI,IXI259,59.78,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI259-HH-1804-T1.nii.gz
+226,IXI,IXI260,60.91,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI260-HH-1805-T1.nii.gz
+227,IXI,IXI261,34.21,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI261-HH-1704-T1.nii.gz
+228,IXI,IXI262,20.93,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI262-HH-1861-T1.nii.gz
+229,IXI,IXI263,25.21,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI263-HH-1684-T1.nii.gz
+230,IXI,IXI264,44.88,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI264-Guys-0854-T1.nii.gz
+231,IXI,IXI265,27.38,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI265-Guys-0845-T1.nii.gz
+232,IXI,IXI266,61.07,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI266-Guys-0853-T1.nii.gz
+233,IXI,IXI267,35.45,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI267-HH-1772-T1.nii.gz
+234,IXI,IXI268,44.92,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI268-Guys-0858-T1.nii.gz
+235,IXI,IXI269,26.58,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI269-Guys-0839-T1.nii.gz
+236,IXI,IXI270,61.02,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI270-Guys-0847-T1.nii.gz
+237,IXI,IXI274,73.54,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI274-HH-2294-T1.nii.gz
+238,IXI,IXI275,24.18,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI275-HH-1803-T1.nii.gz
+239,IXI,IXI276,20.91,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI276-HH-1840-T1.nii.gz
+240,IXI,IXI277,61.34,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI277-HH-1770-T1.nii.gz
+241,IXI,IXI278,20.21,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI278-HH-1771-T1.nii.gz
+242,IXI,IXI279,51.88,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI279-Guys-1044-T1.nii.gz
+243,IXI,IXI280,42.51,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI280-HH-1860-T1.nii.gz
+244,IXI,IXI282,40.01,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI282-HH-2025-T1.nii.gz
+245,IXI,IXI284,63.04,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI284-HH-2354-T1.nii.gz
+246,IXI,IXI285,40.81,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI285-Guys-0857-T1.nii.gz
+247,IXI,IXI286,54.28,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI286-Guys-0859-T1.nii.gz
+248,IXI,IXI287,44.99,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI287-Guys-0863-T1.nii.gz
+249,IXI,IXI288,78.58,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI288-Guys-0879-T1.nii.gz
+250,IXI,IXI289,60.84,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI289-Guys-0864-T1.nii.gz
+251,IXI,IXI290,45.75,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI290-IOP-0874-T1.nii.gz
+252,IXI,IXI291,30.69,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI291-IOP-0882-T1.nii.gz
+253,IXI,IXI292,23.73,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI292-IOP-0877-T1.nii.gz
+254,IXI,IXI293,26.52,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI293-IOP-0876-T1.nii.gz
+255,IXI,IXI294,27.08,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI294-IOP-0868-T1.nii.gz
+256,IXI,IXI295,50.9,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI295-HH-1814-T1.nii.gz
+257,IXI,IXI296,63.97,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI296-HH-1970-T1.nii.gz
+258,IXI,IXI297,51.29,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI297-Guys-0886-T1.nii.gz
+259,IXI,IXI298,43.46,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI298-Guys-0861-T1.nii.gz
+260,IXI,IXI299,53.41,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI299-Guys-0893-T1.nii.gz
+261,IXI,IXI300,38.28,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI300-Guys-0880-T1.nii.gz
+262,IXI,IXI302,59.51,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI302-HH-1883-T1.nii.gz
+263,IXI,IXI303,25.46,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI303-IOP-0968-T1.nii.gz
+264,IXI,IXI304,46.59,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI304-Guys-0862-T1.nii.gz
+265,IXI,IXI305,62.51,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI305-IOP-0871-T1.nii.gz
+266,IXI,IXI306,30.18,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI306-IOP-0867-T1.nii.gz
+267,IXI,IXI307,33.0,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI307-IOP-0872-T1.nii.gz
+268,IXI,IXI308,41.2,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI308-Guys-0884-T1.nii.gz
+269,IXI,IXI309,33.92,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI309-IOP-0897-T1.nii.gz
+270,IXI,IXI310,62.02,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI310-IOP-0890-T1.nii.gz
+271,IXI,IXI311,61.59,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI311-Guys-0885-T1.nii.gz
+272,IXI,IXI312,59.15,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI312-Guys-0887-T1.nii.gz
+273,IXI,IXI313,65.85,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI313-HH-2241-T1.nii.gz
+274,IXI,IXI314,45.86,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI314-IOP-0889-T1.nii.gz
+275,IXI,IXI315,25.45,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI315-IOP-0888-T1.nii.gz
+276,IXI,IXI316,56.94,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI316-HH-1862-T1.nii.gz
+277,IXI,IXI317,67.79,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI317-Guys-0896-T1.nii.gz
+278,IXI,IXI318,54.49,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI318-Guys-0895-T1.nii.gz
+279,IXI,IXI319,59.47,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI319-Guys-0901-T1.nii.gz
+280,IXI,IXI320,29.74,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI320-Guys-0902-T1.nii.gz
+281,IXI,IXI321,54.74,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI321-Guys-0903-T1.nii.gz
+282,IXI,IXI322,28.48,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI322-IOP-0891-T1.nii.gz
+283,IXI,IXI324,60.08,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI324-Guys-0922-T1.nii.gz
+284,IXI,IXI325,52.67,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI325-Guys-0911-T1.nii.gz
+285,IXI,IXI326,67.74,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI326-Guys-0907-T1.nii.gz
+286,IXI,IXI327,73.56,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI327-HH-1999-T1.nii.gz
+287,IXI,IXI328,61.79,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI328-HH-2295-T1.nii.gz
+288,IXI,IXI328,62.08,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI328-HH-2295-T1.nii.gz
+289,IXI,IXI329,55.66,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI329-HH-1908-T1.nii.gz
+290,IXI,IXI330,44.48,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI330-Guys-0881-T1.nii.gz
+291,IXI,IXI331,23.49,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI331-IOP-0892-T1.nii.gz
+292,IXI,IXI332,42.37,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI332-IOP-1134-T1.nii.gz
+293,IXI,IXI334,69.29,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI334-HH-1907-T1.nii.gz
+294,IXI,IXI335,69.28,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI335-HH-1906-T1.nii.gz
+295,IXI,IXI336,67.77,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI336-Guys-0904-T1.nii.gz
+296,IXI,IXI338,61.49,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI338-HH-1971-T1.nii.gz
+297,IXI,IXI342,48.68,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI342-Guys-0909-T1.nii.gz
+298,IXI,IXI344,62.09,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI344-Guys-0905-T1.nii.gz
+299,IXI,IXI348,38.59,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI348-Guys-0910-T1.nii.gz
+300,IXI,IXI350,62.68,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI350-Guys-0908-T1.nii.gz
+301,IXI,IXI351,56.92,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI351-Guys-0914-T1.nii.gz
+302,IXI,IXI353,52.69,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI353-HH-1996-T1.nii.gz
+303,IXI,IXI354,45.62,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI354-HH-2024-T1.nii.gz
+304,IXI,IXI356,64.1,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI356-HH-2049-T1.nii.gz
+305,IXI,IXI357,63.54,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI357-HH-2076-T1.nii.gz
+306,IXI,IXI358,62.06,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI358-Guys-0919-T1.nii.gz
+307,IXI,IXI359,62.06,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI359-Guys-0918-T1.nii.gz
+308,IXI,IXI360,54.19,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI360-Guys-0912-T1.nii.gz
+309,IXI,IXI361,54.19,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI361-Guys-0913-T1.nii.gz
+310,IXI,IXI362,41.1,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI362-HH-2051-T1.nii.gz
+311,IXI,IXI363,60.92,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI363-HH-2050-T1.nii.gz
+312,IXI,IXI364,67.16,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI364-Guys-0920-T1.nii.gz
+313,IXI,IXI365,66.53,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI365-Guys-0923-T1.nii.gz
+314,IXI,IXI367,54.21,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI367-Guys-0917-T1.nii.gz
+315,IXI,IXI368,54.21,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI368-Guys-0916-T1.nii.gz
+316,IXI,IXI369,60.6,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI369-Guys-0924-T1.nii.gz
+317,IXI,IXI370,60.6,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI370-Guys-0921-T1.nii.gz
+318,IXI,IXI371,26.4,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI371-IOP-0970-T1.nii.gz
+319,IXI,IXI372,29.26,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI372-IOP-0971-T1.nii.gz
+320,IXI,IXI373,58.79,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI373-IOP-0967-T1.nii.gz
+321,IXI,IXI375,41.06,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI375-Guys-0925-T1.nii.gz
+322,IXI,IXI376,72.59,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI376-Guys-0938-T1.nii.gz
+323,IXI,IXI377,72.59,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI377-Guys-0937-T1.nii.gz
+324,IXI,IXI378,38.84,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI378-IOP-0972-T1.nii.gz
+325,IXI,IXI379,66.91,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI379-Guys-0943-T1.nii.gz
+326,IXI,IXI380,66.91,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI380-Guys-0944-T1.nii.gz
+327,IXI,IXI381,59.67,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI381-Guys-1024-T1.nii.gz
+328,IXI,IXI382,65.91,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI382-IOP-1135-T1.nii.gz
+329,IXI,IXI383,74.01,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI383-HH-2099-T1.nii.gz
+330,IXI,IXI384,74.01,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI384-HH-2100-T1.nii.gz
+331,IXI,IXI385,64.65,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI385-HH-2078-T1.nii.gz
+332,IXI,IXI386,52.3,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI386-HH-2077-T1.nii.gz
+333,IXI,IXI387,46.43,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI387-HH-2101-T1.nii.gz
+334,IXI,IXI388,33.35,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI388-IOP-0973-T1.nii.gz
+335,IXI,IXI389,58.37,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI389-Guys-0930-T1.nii.gz
+336,IXI,IXI390,41.89,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI390-Guys-0931-T1.nii.gz
+337,IXI,IXI391,52.53,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI391-Guys-0934-T1.nii.gz
+338,IXI,IXI392,62.46,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI392-Guys-1064-T1.nii.gz
+339,IXI,IXI393,60.1,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI393-Guys-0941-T1.nii.gz
+340,IXI,IXI394,53.75,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI394-Guys-0940-T1.nii.gz
+341,IXI,IXI395,45.03,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI395-IOP-0969-T1.nii.gz
+342,IXI,IXI396,53.57,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI396-HH-2115-T1.nii.gz
+343,IXI,IXI397,51.66,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI397-Guys-0953-T1.nii.gz
+344,IXI,IXI398,51.66,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI398-Guys-0952-T1.nii.gz
+345,IXI,IXI399,73.54,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI399-Guys-0966-T1.nii.gz
+346,IXI,IXI400,55.58,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI400-Guys-0977-T1.nii.gz
+347,IXI,IXI401,55.58,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI401-Guys-0978-T1.nii.gz
+348,IXI,IXI402,53.54,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI402-Guys-0961-T1.nii.gz
+349,IXI,IXI403,53.54,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI403-Guys-0965-T1.nii.gz
+350,IXI,IXI404,62.74,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI404-Guys-0950-T1.nii.gz
+351,IXI,IXI405,58.57,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI405-Guys-0948-T1.nii.gz
+352,IXI,IXI406,58.28,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI406-Guys-0963-T1.nii.gz
+353,IXI,IXI407,58.28,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI407-Guys-0964-T1.nii.gz
+354,IXI,IXI408,52.13,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI408-Guys-0962-T1.nii.gz
+355,IXI,IXI409,70.95,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI409-Guys-0960-T1.nii.gz
+356,IXI,IXI410,53.41,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI410-Guys-0958-T1.nii.gz
+357,IXI,IXI411,53.41,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI411-Guys-0959-T1.nii.gz
+358,IXI,IXI412,58.57,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI412-Guys-0949-T1.nii.gz
+359,IXI,IXI413,59.84,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI413-Guys-0955-T1.nii.gz
+360,IXI,IXI414,59.74,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI414-Guys-0957-T1.nii.gz
+361,IXI,IXI415,72.33,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI415-Guys-0942-T1.nii.gz
+362,IXI,IXI416,72.15,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI416-Guys-1051-T1.nii.gz
+363,IXI,IXI417,59.88,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI417-Guys-0939-T1.nii.gz
+364,IXI,IXI418,59.74,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI418-Guys-0956-T1.nii.gz
+365,IXI,IXI419,73.18,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI419-Guys-1027-T1.nii.gz
+366,IXI,IXI420,73.18,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI420-Guys-1028-T1.nii.gz
+367,IXI,IXI422,37.1,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI422-Guys-1071-T1.nii.gz
+368,IXI,IXI423,39.02,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI423-IOP-0974-T1.nii.gz
+369,IXI,IXI424,46.3,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI424-IOP-0991-T1.nii.gz
+370,IXI,IXI425,19.98,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI425-IOP-0988-T1.nii.gz
+371,IXI,IXI426,23.11,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI426-IOP-1011-T1.nii.gz
+372,IXI,IXI427,34.28,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI427-IOP-1012-T1.nii.gz
+373,IXI,IXI428,55.54,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI428-Guys-0996-T1.nii.gz
+374,IXI,IXI429,55.54,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI429-Guys-0997-T1.nii.gz
+375,IXI,IXI430,72.01,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI430-IOP-0990-T1.nii.gz
+376,IXI,IXI431,58.78,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI431-Guys-0986-T1.nii.gz
+377,IXI,IXI432,57.74,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI432-Guys-0987-T1.nii.gz
+378,IXI,IXI433,73.9,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI433-IOP-0989-T1.nii.gz
+379,IXI,IXI434,67.24,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI434-IOP-1010-T1.nii.gz
+380,IXI,IXI436,28.35,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI436-HH-2153-T1.nii.gz
+381,IXI,IXI437,55.6,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI437-HH-2152-T1.nii.gz
+382,IXI,IXI438,63.68,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI438-HH-2155-T1.nii.gz
+383,IXI,IXI439,43.92,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI439-HH-2114-T1.nii.gz
+384,IXI,IXI440,48.1,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI440-HH-2127-T1.nii.gz
+385,IXI,IXI441,48.1,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI441-HH-2154-T1.nii.gz
+386,IXI,IXI442,61.37,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI442-IOP-1041-T1.nii.gz
+387,IXI,IXI443,69.13,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI443-HH-2215-T1.nii.gz
+388,IXI,IXI444,58.38,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI444-HH-2270-T1.nii.gz
+389,IXI,IXI445,60.03,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI445-HH-2269-T1.nii.gz
+390,IXI,IXI446,60.11,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI446-HH-2325-T1.nii.gz
+391,IXI,IXI447,72.83,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI447-Guys-0979-T1.nii.gz
+392,IXI,IXI448,79.41,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI448-HH-2393-T1.nii.gz
+393,IXI,IXI449,46.9,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI449-Guys-1082-T1.nii.gz
+394,IXI,IXI450,68.51,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI450-Guys-1093-T1.nii.gz
+395,IXI,IXI451,41.13,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI451-HH-2212-T1.nii.gz
+396,IXI,IXI452,63.4,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI452-HH-2213-T1.nii.gz
+397,IXI,IXI453,63.42,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI453-HH-2214-T1.nii.gz
+398,IXI,IXI454,32.58,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI454-Guys-1083-T1.nii.gz
+399,IXI,IXI455,68.48,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI455-Guys-0981-T1.nii.gz
+400,IXI,IXI456,70.25,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI456-Guys-1019-T1.nii.gz
+401,IXI,IXI458,70.71,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI458-Guys-0993-T1.nii.gz
+402,IXI,IXI459,62.43,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI459-Guys-0992-T1.nii.gz
+403,IXI,IXI460,60.9,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI460-Guys-0999-T1.nii.gz
+404,IXI,IXI461,60.9,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI461-Guys-0998-T1.nii.gz
+405,IXI,IXI462,75.08,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI462-IOP-1042-T1.nii.gz
+406,IXI,IXI463,70.14,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI463-IOP-1043-T1.nii.gz
+407,IXI,IXI464,86.32,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI464-IOP-1029-T1.nii.gz
+408,IXI,IXI465,69.28,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI465-HH-2176-T1.nii.gz
+409,IXI,IXI467,66.95,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI467-Guys-0983-T1.nii.gz
+410,IXI,IXI468,67.7,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI468-Guys-0985-T1.nii.gz
+411,IXI,IXI469,32.84,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI469-IOP-1136-T1.nii.gz
+412,IXI,IXI470,35.98,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI470-IOP-1030-T1.nii.gz
+413,IXI,IXI473,68.13,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI473-IOP-1137-T1.nii.gz
+414,IXI,IXI474,34.01,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI474-IOP-1138-T1.nii.gz
+415,IXI,IXI475,32.7,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI475-IOP-1139-T1.nii.gz
+416,IXI,IXI476,78.36,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI476-IOP-1140-T1.nii.gz
+417,IXI,IXI477,46.43,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI477-IOP-1141-T1.nii.gz
+418,IXI,IXI478,49.02,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI478-IOP-1142-T1.nii.gz
+419,IXI,IXI479,60.02,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI479-Guys-1034-T1.nii.gz
+420,IXI,IXI480,60.02,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI480-Guys-1033-T1.nii.gz
+421,IXI,IXI481,65.33,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI481-HH-2175-T1.nii.gz
+422,IXI,IXI482,72.09,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI482-HH-2178-T1.nii.gz
+423,IXI,IXI483,63.39,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI483-HH-2177-T1.nii.gz
+424,IXI,IXI484,66.37,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI484-HH-2179-T1.nii.gz
+425,IXI,IXI485,68.85,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI485-HH-2180-T1.nii.gz
+426,IXI,IXI486,63.84,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI486-Guys-1005-T1.nii.gz
+427,IXI,IXI487,66.28,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI487-Guys-1037-T1.nii.gz
+428,IXI,IXI488,69.07,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI488-Guys-1015-T1.nii.gz
+429,IXI,IXI489,69.07,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI489-Guys-1014-T1.nii.gz
+430,IXI,IXI490,70.14,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI490-Guys-1018-T1.nii.gz
+431,IXI,IXI491,75.97,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI491-Guys-1032-T1.nii.gz
+432,IXI,IXI492,69.29,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI492-Guys-1022-T1.nii.gz
+433,IXI,IXI493,67.77,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI493-Guys-1007-T1.nii.gz
+434,IXI,IXI494,66.33,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI494-Guys-1008-T1.nii.gz
+435,IXI,IXI495,60.87,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI495-Guys-1009-T1.nii.gz
+436,IXI,IXI496,62.8,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI496-Guys-1045-T1.nii.gz
+437,IXI,IXI497,72.31,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI497-Guys-1002-T1.nii.gz
+438,IXI,IXI498,76.78,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI498-Guys-1050-T1.nii.gz
+439,IXI,IXI499,82.19,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI499-Guys-1004-T1.nii.gz
+440,IXI,IXI500,63.18,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI500-Guys-1017-T1.nii.gz
+441,IXI,IXI501,67.11,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI501-Guys-1016-T1.nii.gz
+442,IXI,IXI502,63.88,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI502-Guys-1020-T1.nii.gz
+443,IXI,IXI503,65.1,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI503-Guys-1021-T1.nii.gz
+444,IXI,IXI504,68.15,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI504-Guys-1025-T1.nii.gz
+445,IXI,IXI505,62.25,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI505-Guys-1026-T1.nii.gz
+446,IXI,IXI506,69.69,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI506-Guys-1035-T1.nii.gz
+447,IXI,IXI507,71.66,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI507-Guys-1036-T1.nii.gz
+448,IXI,IXI508,60.93,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI508-HH-2268-T1.nii.gz
+449,IXI,IXI510,32.42,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI510-IOP-1143-T1.nii.gz
+450,IXI,IXI511,64.58,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI511-HH-2238-T1.nii.gz
+451,IXI,IXI512,64.75,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI512-Guys-1054-T1.nii.gz
+452,IXI,IXI515,62.09,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI515-HH-2377-T1.nii.gz
+453,IXI,IXI516,60.41,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI516-HH-2297-T1.nii.gz
+454,IXI,IXI517,36.16,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI517-IOP-1144-T1.nii.gz
+455,IXI,IXI518,75.83,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI518-HH-2239-T1.nii.gz
+456,IXI,IXI519,57.51,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI519-HH-2240-T1.nii.gz
+457,IXI,IXI521,70.84,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI521-HH-2353-T1.nii.gz
+458,IXI,IXI522,28.93,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI522-HH-2453-T1.nii.gz
+459,IXI,IXI523,63.42,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI523-Guys-1056-T1.nii.gz
+460,IXI,IXI524,33.52,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI524-HH-2412-T1.nii.gz
+461,IXI,IXI525,41.52,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI525-HH-2413-T1.nii.gz
+462,IXI,IXI526,54.61,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI526-HH-2392-T1.nii.gz
+463,IXI,IXI527,53.33,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI527-HH-2376-T1.nii.gz
+464,IXI,IXI528,40.53,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI528-Guys-1073-T1.nii.gz
+465,IXI,IXI531,75.94,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI531-Guys-1057-T1.nii.gz
+466,IXI,IXI532,36.55,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI532-IOP-1145-T1.nii.gz
+467,IXI,IXI533,73.6,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI533-Guys-1066-T1.nii.gz
+468,IXI,IXI534,69.61,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI534-Guys-1062-T1.nii.gz
+469,IXI,IXI535,65.26,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI535-Guys-1061-T1.nii.gz
+470,IXI,IXI536,58.83,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI536-Guys-1059-T1.nii.gz
+471,IXI,IXI537,67.46,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI537-HH-2378-T1.nii.gz
+472,IXI,IXI538,81.94,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI538-HH-2411-T1.nii.gz
+473,IXI,IXI539,78.07,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI539-Guys-1067-T1.nii.gz
+474,IXI,IXI541,36.42,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI541-IOP-1146-T1.nii.gz
+475,IXI,IXI542,44.38,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI542-IOP-1147-T1.nii.gz
+476,IXI,IXI543,70.52,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI543-IOP-1148-T1.nii.gz
+477,IXI,IXI544,58.04,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI544-HH-2395-T1.nii.gz
+478,IXI,IXI546,60.07,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI546-HH-2450-T1.nii.gz
+479,IXI,IXI547,33.83,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI547-IOP-1149-T1.nii.gz
+480,IXI,IXI548,66.33,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI548-IOP-1150-T1.nii.gz
+481,IXI,IXI549,25.02,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI549-Guys-1046-T1.nii.gz
+482,IXI,IXI550,59.91,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI550-Guys-1069-T1.nii.gz
+483,IXI,IXI551,70.5,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI551-Guys-1065-T1.nii.gz
+484,IXI,IXI552,64.66,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI552-Guys-1063-T1.nii.gz
+485,IXI,IXI553,28.24,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI553-IOP-1151-T1.nii.gz
+486,IXI,IXI554,70.11,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI554-Guys-1068-T1.nii.gz
+487,IXI,IXI555,65.67,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI555-Guys-1074-T1.nii.gz
+488,IXI,IXI556,55.64,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI556-HH-2452-T1.nii.gz
+489,IXI,IXI558,66.86,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI558-Guys-1079-T1.nii.gz
+490,IXI,IXI559,67.82,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI559-HH-2394-T1.nii.gz
+491,IXI,IXI560,63.87,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI560-Guys-1070-T1.nii.gz
+492,IXI,IXI561,31.58,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI561-IOP-1152-T1.nii.gz
+493,IXI,IXI562,42.79,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI562-Guys-1131-T1.nii.gz
+494,IXI,IXI563,28.37,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI563-IOP-1153-T1.nii.gz
+495,IXI,IXI565,63.82,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI565-HH-2534-T1.nii.gz
+496,IXI,IXI566,42.97,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI566-HH-2535-T1.nii.gz
+497,IXI,IXI567,28.56,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI567-HH-2536-T1.nii.gz
+498,IXI,IXI568,33.69,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI568-HH-2607-T1.nii.gz
+499,IXI,IXI569,56.96,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI569-Guys-1101-T1.nii.gz
+500,IXI,IXI571,38.89,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI571-IOP-1154-T1.nii.gz
+501,IXI,IXI572,41.95,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI572-HH-2605-T1.nii.gz
+502,IXI,IXI573,35.46,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI573-IOP-1155-T1.nii.gz
+503,IXI,IXI574,50.57,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI574-IOP-1156-T1.nii.gz
+504,IXI,IXI575,57.75,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI575-HH-2658-T1.nii.gz
+505,IXI,IXI576,67.33,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI576-Guys-1077-T1.nii.gz
+506,IXI,IXI577,64.19,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI577-HH-2661-T1.nii.gz
+507,IXI,IXI578,57.86,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI578-Guys-1078-T1.nii.gz
+508,IXI,IXI579,41.22,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI579-Guys-1126-T1.nii.gz
+509,IXI,IXI582,29.77,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI582-Guys-1127-T1.nii.gz
+510,IXI,IXI584,41.33,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI584-Guys-1129-T1.nii.gz
+511,IXI,IXI585,28.12,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI585-Guys-1130-T1.nii.gz
+512,IXI,IXI586,34.37,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI586-HH-2451-T1.nii.gz
+513,IXI,IXI587,33.59,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI587-Guys-1128-T1.nii.gz
+514,IXI,IXI588,45.3,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI588-IOP-1158-T1.nii.gz
+515,IXI,IXI591,59.89,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI591-Guys-1084-T1.nii.gz
+516,IXI,IXI592,48.08,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI592-Guys-1085-T1.nii.gz
+517,IXI,IXI593,55.22,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI593-Guys-1109-T1.nii.gz
+518,IXI,IXI594,62.0,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI594-Guys-1089-T1.nii.gz
+519,IXI,IXI595,49.6,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI595-IOP-1159-T1.nii.gz
+520,IXI,IXI596,29.66,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI596-IOP-1160-T1.nii.gz
+521,IXI,IXI597,43.08,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI597-IOP-1161-T1.nii.gz
+522,IXI,IXI598,52.89,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI598-HH-2606-T1.nii.gz
+523,IXI,IXI599,39.38,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI599-HH-2659-T1.nii.gz
+524,IXI,IXI600,38.96,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI600-HH-2660-T1.nii.gz
+525,IXI,IXI601,35.52,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI601-HH-2700-T1.nii.gz
+526,IXI,IXI603,66.02,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI603-HH-2701-T1.nii.gz
+527,IXI,IXI605,70.33,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI605-HH-2598-T1.nii.gz
+528,IXI,IXI606,60.54,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI606-HH-2601-T1.nii.gz
+529,IXI,IXI607,83.81,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI607-Guys-1097-T1.nii.gz
+530,IXI,IXI608,52.5,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI608-HH-2599-T1.nii.gz
+531,IXI,IXI609,29.0,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI609-HH-2600-T1.nii.gz
+532,IXI,IXI610,56.68,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI610-HH-2649-T1.nii.gz
+533,IXI,IXI611,28.87,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI611-HH-2650-T1.nii.gz
+534,IXI,IXI612,33.92,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI612-HH-2688-T1.nii.gz
+535,IXI,IXI613,25.59,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI613-HH-2734-T1.nii.gz
+536,IXI,IXI614,26.93,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI614-HH-2735-T1.nii.gz
+537,IXI,IXI616,55.09,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI616-Guys-1092-T1.nii.gz
+538,IXI,IXI617,39.18,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI617-Guys-1090-T1.nii.gz
+539,IXI,IXI618,37.95,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI618-Guys-1091-T1.nii.gz
+540,IXI,IXI619,24.88,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI619-Guys-1099-T1.nii.gz
+541,IXI,IXI621,31.62,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI621-Guys-1100-T1.nii.gz
+542,IXI,IXI622,59.69,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI622-Guys-1102-T1.nii.gz
+543,IXI,IXI625,47.07,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI625-Guys-1098-T1.nii.gz
+544,IXI,IXI626,64.27,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI626-Guys-1094-T1.nii.gz
+545,IXI,IXI627,68.48,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI627-Guys-1103-T1.nii.gz
+546,IXI,IXI629,59.26,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI629-Guys-1095-T1.nii.gz
+547,IXI,IXI630,54.69,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI630-Guys-1108-T1.nii.gz
+548,IXI,IXI631,41.3,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI631-HH-2651-T1.nii.gz
+549,IXI,IXI632,56.44,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI632-HH-2652-T1.nii.gz
+550,IXI,IXI633,36.71,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI633-HH-2689-T1.nii.gz
+551,IXI,IXI634,34.06,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI634-HH-2690-T1.nii.gz
+552,IXI,IXI635,25.53,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI635-HH-2691-T1.nii.gz
+553,IXI,IXI636,43.16,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI636-HH-2733-T1.nii.gz
+554,IXI,IXI639,86.2,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI639-Guys-1088-T1.nii.gz
+555,IXI,IXI640,80.8,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI640-Guys-1106-T1.nii.gz
+556,IXI,IXI641,68.12,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI641-Guys-1105-T1.nii.gz
+557,IXI,IXI642,57.07,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI642-Guys-1104-T1.nii.gz
+558,IXI,IXI644,56.87,test,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI644-Guys-1121-T1.nii.gz
+559,IXI,IXI646,71.21,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI646-HH-2653-T1.nii.gz
+560,IXI,IXI648,47.72,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI648-Guys-1107-T1.nii.gz
+561,IXI,IXI651,50.4,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI651-Guys-1118-T1.nii.gz
+562,IXI,IXI652,42.99,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI652-Guys-1116-T1.nii.gz
+563,IXI,IXI653,46.22,train,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI653-Guys-1122-T1.nii.gz
+564,IXI,IXI662,41.74,dev,/home/nafiu/BasicData/cse472_DL_project/ixi_storage/IXI-T1/IXI662-Guys-1120-T1.nii.gz
+565,FCON1000,fcon1000AnnArbor_sub00306,68.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub00306.nii.gz
+566,FCON1000,fcon1000AnnArbor_sub04111,25.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub04111.nii.gz
+567,FCON1000,fcon1000AnnArbor_sub04619,18.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub04619.nii.gz
+568,FCON1000,fcon1000AnnArbor_sub05580,77.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub05580.nii.gz
+569,FCON1000,fcon1000AnnArbor_sub07921,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub07921.nii.gz
+570,FCON1000,fcon1000AnnArbor_sub11043,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub11043.nii.gz
+571,FCON1000,fcon1000AnnArbor_sub13636,29.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub13636.nii.gz
+572,FCON1000,fcon1000AnnArbor_sub13959,31.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub13959.nii.gz
+573,FCON1000,fcon1000AnnArbor_sub15846,70.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub15846.nii.gz
+574,FCON1000,fcon1000AnnArbor_sub16960,13.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub16960.nii.gz
+575,FCON1000,fcon1000AnnArbor_sub18546,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub18546.nii.gz
+576,FCON1000,fcon1000AnnArbor_sub18698,29.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub18698.nii.gz
+577,FCON1000,fcon1000AnnArbor_sub20317,15.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub20317.nii.gz
+578,FCON1000,fcon1000AnnArbor_sub26099,73.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub26099.nii.gz
+579,FCON1000,fcon1000AnnArbor_sub28433,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub28433.nii.gz
+580,FCON1000,fcon1000AnnArbor_sub30250,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub30250.nii.gz
+581,FCON1000,fcon1000AnnArbor_sub30421,17.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub30421.nii.gz
+582,FCON1000,fcon1000AnnArbor_sub33437,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub33437.nii.gz
+583,FCON1000,fcon1000AnnArbor_sub34781,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub34781.nii.gz
+584,FCON1000,fcon1000AnnArbor_sub38614,14.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub38614.nii.gz
+585,FCON1000,fcon1000AnnArbor_sub39635,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub39635.nii.gz
+586,FCON1000,fcon1000AnnArbor_sub39923,70.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub39923.nii.gz
+587,FCON1000,fcon1000AnnArbor_sub42616,21.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub42616.nii.gz
+588,FCON1000,fcon1000AnnArbor_sub43409,71.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub43409.nii.gz
+589,FCON1000,fcon1000AnnArbor_sub45569,63.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub45569.nii.gz
+590,FCON1000,fcon1000AnnArbor_sub45660,16.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub45660.nii.gz
+591,FCON1000,fcon1000AnnArbor_sub46727,13.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub46727.nii.gz
+592,FCON1000,fcon1000AnnArbor_sub47191,27.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub47191.nii.gz
+593,FCON1000,fcon1000AnnArbor_sub47659,33.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub47659.nii.gz
+594,FCON1000,fcon1000AnnArbor_sub49687,13.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub49687.nii.gz
+595,FCON1000,fcon1000AnnArbor_sub50859,79.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub50859.nii.gz
+596,FCON1000,fcon1000AnnArbor_sub51248,40.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub51248.nii.gz
+597,FCON1000,fcon1000AnnArbor_sub53269,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub53269.nii.gz
+598,FCON1000,fcon1000AnnArbor_sub53959,20.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub53959.nii.gz
+599,FCON1000,fcon1000AnnArbor_sub56028,26.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub56028.nii.gz
+600,FCON1000,fcon1000AnnArbor_sub56686,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub56686.nii.gz
+601,FCON1000,fcon1000AnnArbor_sub57025,68.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub57025.nii.gz
+602,FCON1000,fcon1000AnnArbor_sub57196,77.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub57196.nii.gz
+603,FCON1000,fcon1000AnnArbor_sub59573,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub59573.nii.gz
+604,FCON1000,fcon1000AnnArbor_sub62819,65.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub62819.nii.gz
+605,FCON1000,fcon1000AnnArbor_sub64831,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub64831.nii.gz
+606,FCON1000,fcon1000AnnArbor_sub64969,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub64969.nii.gz
+607,FCON1000,fcon1000AnnArbor_sub66506,78.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub66506.nii.gz
+608,FCON1000,fcon1000AnnArbor_sub70106,17.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub70106.nii.gz
+609,FCON1000,fcon1000AnnArbor_sub72215,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub72215.nii.gz
+610,FCON1000,fcon1000AnnArbor_sub72416,80.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub72416.nii.gz
+611,FCON1000,fcon1000AnnArbor_sub73168,78.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub73168.nii.gz
+612,FCON1000,fcon1000AnnArbor_sub73812,28.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub73812.nii.gz
+613,FCON1000,fcon1000AnnArbor_sub75922,18.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub75922.nii.gz
+614,FCON1000,fcon1000AnnArbor_sub78151,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub78151.nii.gz
+615,FCON1000,fcon1000AnnArbor_sub82334,14.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub82334.nii.gz
+616,FCON1000,fcon1000AnnArbor_sub85257,26.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub85257.nii.gz
+617,FCON1000,fcon1000AnnArbor_sub86367,15.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub86367.nii.gz
+618,FCON1000,fcon1000AnnArbor_sub87745,14.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub87745.nii.gz
+619,FCON1000,fcon1000AnnArbor_sub90127,77.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub90127.nii.gz
+620,FCON1000,fcon1000AnnArbor_sub90950,66.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub90950.nii.gz
+621,FCON1000,fcon1000AnnArbor_sub96621,15.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub96621.nii.gz
+622,FCON1000,fcon1000AnnArbor_sub97518,21.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub97518.nii.gz
+623,FCON1000,fcon1000AnnArbor_sub98007,66.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub98007.nii.gz
+624,FCON1000,fcon1000AnnArbor_sub99692,66.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/AnnArbor_sub99692.nii.gz
+625,FCON1000,fcon1000Atlanta_sub00354,28.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub00354.nii.gz
+626,FCON1000,fcon1000Atlanta_sub00368,28.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub00368.nii.gz
+627,FCON1000,fcon1000Atlanta_sub06870,57.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub06870.nii.gz
+628,FCON1000,fcon1000Atlanta_sub07145,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub07145.nii.gz
+629,FCON1000,fcon1000Atlanta_sub15817,31.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub15817.nii.gz
+630,FCON1000,fcon1000Atlanta_sub16563,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub16563.nii.gz
+631,FCON1000,fcon1000Atlanta_sub18219,33.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub18219.nii.gz
+632,FCON1000,fcon1000Atlanta_sub18702,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub18702.nii.gz
+633,FCON1000,fcon1000Atlanta_sub24972,24.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub24972.nii.gz
+634,FCON1000,fcon1000Atlanta_sub26938,31.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub26938.nii.gz
+635,FCON1000,fcon1000Atlanta_sub32093,27.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub32093.nii.gz
+636,FCON1000,fcon1000Atlanta_sub49816,27.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub49816.nii.gz
+637,FCON1000,fcon1000Atlanta_sub52783,,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub52783.nii.gz
+638,FCON1000,fcon1000Atlanta_sub53122,28.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub53122.nii.gz
+639,FCON1000,fcon1000Atlanta_sub55652,25.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub55652.nii.gz
+640,FCON1000,fcon1000Atlanta_sub58250,54.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub58250.nii.gz
+641,FCON1000,fcon1000Atlanta_sub59806,23.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub59806.nii.gz
+642,FCON1000,fcon1000Atlanta_sub60499,29.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub60499.nii.gz
+643,FCON1000,fcon1000Atlanta_sub61442,29.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub61442.nii.gz
+644,FCON1000,fcon1000Atlanta_sub61902,23.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub61902.nii.gz
+645,FCON1000,fcon1000Atlanta_sub71337,27.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub71337.nii.gz
+646,FCON1000,fcon1000Atlanta_sub72096,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub72096.nii.gz
+647,FCON1000,fcon1000Atlanta_sub72971,30.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub72971.nii.gz
+648,FCON1000,fcon1000Atlanta_sub75153,32.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub75153.nii.gz
+649,FCON1000,fcon1000Atlanta_sub76280,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub76280.nii.gz
+650,FCON1000,fcon1000Atlanta_sub81596,53.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub81596.nii.gz
+651,FCON1000,fcon1000Atlanta_sub86323,31.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub86323.nii.gz
+652,FCON1000,fcon1000Atlanta_sub91049,48.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Atlanta_sub91049.nii.gz
+653,FCON1000,fcon1000Baltimore_sub17017,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub17017.nii.gz
+654,FCON1000,fcon1000Baltimore_sub19738,27.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub19738.nii.gz
+655,FCON1000,fcon1000Baltimore_sub23750,37.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub23750.nii.gz
+656,FCON1000,fcon1000Baltimore_sub23927,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub23927.nii.gz
+657,FCON1000,fcon1000Baltimore_sub29158,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub29158.nii.gz
+658,FCON1000,fcon1000Baltimore_sub30072,26.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub30072.nii.gz
+659,FCON1000,fcon1000Baltimore_sub31837,34.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub31837.nii.gz
+660,FCON1000,fcon1000Baltimore_sub37548,30.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub37548.nii.gz
+661,FCON1000,fcon1000Baltimore_sub52358,40.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub52358.nii.gz
+662,FCON1000,fcon1000Baltimore_sub54257,29.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub54257.nii.gz
+663,FCON1000,fcon1000Baltimore_sub54329,30.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub54329.nii.gz
+664,FCON1000,fcon1000Baltimore_sub73823,24.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub73823.nii.gz
+665,FCON1000,fcon1000Baltimore_sub76160,34.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub76160.nii.gz
+666,FCON1000,fcon1000Baltimore_sub77572,27.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub77572.nii.gz
+667,FCON1000,fcon1000Baltimore_sub80221,26.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub80221.nii.gz
+668,FCON1000,fcon1000Baltimore_sub81887,32.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub81887.nii.gz
+669,FCON1000,fcon1000Baltimore_sub85922,30.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub85922.nii.gz
+670,FCON1000,fcon1000Baltimore_sub86414,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub86414.nii.gz
+671,FCON1000,fcon1000Baltimore_sub90658,32.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub90658.nii.gz
+672,FCON1000,fcon1000Baltimore_sub90893,39.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub90893.nii.gz
+673,FCON1000,fcon1000Baltimore_sub91622,33.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub91622.nii.gz
+674,FCON1000,fcon1000Baltimore_sub94042,27.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub94042.nii.gz
+675,FCON1000,fcon1000Baltimore_sub96234,30.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Baltimore_sub96234.nii.gz
+676,FCON1000,fcon1000Bangor_sub00031,29.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Bangor_sub00031.nii.gz
+677,FCON1000,fcon1000Bangor_sub01903,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Bangor_sub01903.nii.gz
+678,FCON1000,fcon1000Bangor_sub03557,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Bangor_sub03557.nii.gz
+679,FCON1000,fcon1000Bangor_sub04097,36.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Bangor_sub04097.nii.gz
+680,FCON1000,fcon1000Bangor_sub14388,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Bangor_sub14388.nii.gz
+681,FCON1000,fcon1000Bangor_sub27519,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Bangor_sub27519.nii.gz
+682,FCON1000,fcon1000Bangor_sub36736,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Bangor_sub36736.nii.gz
+683,FCON1000,fcon1000Bangor_sub46870,28.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Bangor_sub46870.nii.gz
+684,FCON1000,fcon1000Bangor_sub48632,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Bangor_sub48632.nii.gz
+685,FCON1000,fcon1000Bangor_sub61418,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Bangor_sub61418.nii.gz
+686,FCON1000,fcon1000Bangor_sub61908,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Bangor_sub61908.nii.gz
+687,FCON1000,fcon1000Bangor_sub63767,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Bangor_sub63767.nii.gz
+688,FCON1000,fcon1000Bangor_sub66585,19.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Bangor_sub66585.nii.gz
+689,FCON1000,fcon1000Bangor_sub68050,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Bangor_sub68050.nii.gz
+690,FCON1000,fcon1000Bangor_sub73082,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Bangor_sub73082.nii.gz
+691,FCON1000,fcon1000Bangor_sub77520,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Bangor_sub77520.nii.gz
+692,FCON1000,fcon1000Bangor_sub81464,38.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Bangor_sub81464.nii.gz
+693,FCON1000,fcon1000Bangor_sub82625,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Bangor_sub82625.nii.gz
+694,FCON1000,fcon1000Bangor_sub87568,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Bangor_sub87568.nii.gz
+695,FCON1000,fcon1000Bangor_sub91556,22.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Bangor_sub91556.nii.gz
+696,FCON1000,fcon1000Beijing_sub00440,24.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub00440.nii.gz
+697,FCON1000,fcon1000Beijing_sub01018,18.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub01018.nii.gz
+698,FCON1000,fcon1000Beijing_sub01244,25.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub01244.nii.gz
+699,FCON1000,fcon1000Beijing_sub02403,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub02403.nii.gz
+700,FCON1000,fcon1000Beijing_sub04050,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub04050.nii.gz
+701,FCON1000,fcon1000Beijing_sub04191,24.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub04191.nii.gz
+702,FCON1000,fcon1000Beijing_sub05267,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub05267.nii.gz
+703,FCON1000,fcon1000Beijing_sub06880,18.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub06880.nii.gz
+704,FCON1000,fcon1000Beijing_sub06899,19.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub06899.nii.gz
+705,FCON1000,fcon1000Beijing_sub07144,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub07144.nii.gz
+706,FCON1000,fcon1000Beijing_sub07716,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub07716.nii.gz
+707,FCON1000,fcon1000Beijing_sub07717,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub07717.nii.gz
+708,FCON1000,fcon1000Beijing_sub08001,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub08001.nii.gz
+709,FCON1000,fcon1000Beijing_sub08251,24.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub08251.nii.gz
+710,FCON1000,fcon1000Beijing_sub08455,25.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub08455.nii.gz
+711,FCON1000,fcon1000Beijing_sub08816,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub08816.nii.gz
+712,FCON1000,fcon1000Beijing_sub08992,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub08992.nii.gz
+713,FCON1000,fcon1000Beijing_sub10186,19.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub10186.nii.gz
+714,FCON1000,fcon1000Beijing_sub10277,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub10277.nii.gz
+715,FCON1000,fcon1000Beijing_sub10869,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub10869.nii.gz
+716,FCON1000,fcon1000Beijing_sub10973,21.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub10973.nii.gz
+717,FCON1000,fcon1000Beijing_sub11072,25.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub11072.nii.gz
+718,FCON1000,fcon1000Beijing_sub11344,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub11344.nii.gz
+719,FCON1000,fcon1000Beijing_sub12220,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub12220.nii.gz
+720,FCON1000,fcon1000Beijing_sub14238,20.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub14238.nii.gz
+721,FCON1000,fcon1000Beijing_sub15441,19.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub15441.nii.gz
+722,FCON1000,fcon1000Beijing_sub16091,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub16091.nii.gz
+723,FCON1000,fcon1000Beijing_sub16943,26.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub16943.nii.gz
+724,FCON1000,fcon1000Beijing_sub17093,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub17093.nii.gz
+725,FCON1000,fcon1000Beijing_sub17159,19.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub17159.nii.gz
+726,FCON1000,fcon1000Beijing_sub17315,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub17315.nii.gz
+727,FCON1000,fcon1000Beijing_sub17586,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub17586.nii.gz
+728,FCON1000,fcon1000Beijing_sub17603,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub17603.nii.gz
+729,FCON1000,fcon1000Beijing_sub17642,24.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub17642.nii.gz
+730,FCON1000,fcon1000Beijing_sub17688,19.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub17688.nii.gz
+731,FCON1000,fcon1000Beijing_sub18326,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub18326.nii.gz
+732,FCON1000,fcon1000Beijing_sub18758,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub18758.nii.gz
+733,FCON1000,fcon1000Beijing_sub18960,23.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub18960.nii.gz
+734,FCON1000,fcon1000Beijing_sub19642,21.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub19642.nii.gz
+735,FCON1000,fcon1000Beijing_sub19974,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub19974.nii.gz
+736,FCON1000,fcon1000Beijing_sub20127,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub20127.nii.gz
+737,FCON1000,fcon1000Beijing_sub20246,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub20246.nii.gz
+738,FCON1000,fcon1000Beijing_sub20765,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub20765.nii.gz
+739,FCON1000,fcon1000Beijing_sub20948,23.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub20948.nii.gz
+740,FCON1000,fcon1000Beijing_sub21115,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub21115.nii.gz
+741,FCON1000,fcon1000Beijing_sub22201,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub22201.nii.gz
+742,FCON1000,fcon1000Beijing_sub22595,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub22595.nii.gz
+743,FCON1000,fcon1000Beijing_sub22661,21.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub22661.nii.gz
+744,FCON1000,fcon1000Beijing_sub22715,19.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub22715.nii.gz
+745,FCON1000,fcon1000Beijing_sub22890,18.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub22890.nii.gz
+746,FCON1000,fcon1000Beijing_sub26713,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub26713.nii.gz
+747,FCON1000,fcon1000Beijing_sub28206,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub28206.nii.gz
+748,FCON1000,fcon1000Beijing_sub28403,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub28403.nii.gz
+749,FCON1000,fcon1000Beijing_sub28698,18.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub28698.nii.gz
+750,FCON1000,fcon1000Beijing_sub28792,20.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub28792.nii.gz
+751,FCON1000,fcon1000Beijing_sub28801,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub28801.nii.gz
+752,FCON1000,fcon1000Beijing_sub28907,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub28907.nii.gz
+753,FCON1000,fcon1000Beijing_sub28965,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub28965.nii.gz
+754,FCON1000,fcon1000Beijing_sub29590,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub29590.nii.gz
+755,FCON1000,fcon1000Beijing_sub29785,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub29785.nii.gz
+756,FCON1000,fcon1000Beijing_sub30272,23.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub30272.nii.gz
+757,FCON1000,fcon1000Beijing_sub30310,22.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub30310.nii.gz
+758,FCON1000,fcon1000Beijing_sub30556,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub30556.nii.gz
+759,FCON1000,fcon1000Beijing_sub30616,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub30616.nii.gz
+760,FCON1000,fcon1000Beijing_sub30988,23.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub30988.nii.gz
+761,FCON1000,fcon1000Beijing_sub31058,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub31058.nii.gz
+762,FCON1000,fcon1000Beijing_sub31729,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub31729.nii.gz
+763,FCON1000,fcon1000Beijing_sub32517,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub32517.nii.gz
+764,FCON1000,fcon1000Beijing_sub32587,20.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub32587.nii.gz
+765,FCON1000,fcon1000Beijing_sub33747,26.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub33747.nii.gz
+766,FCON1000,fcon1000Beijing_sub33943,20.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub33943.nii.gz
+767,FCON1000,fcon1000Beijing_sub33991,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub33991.nii.gz
+768,FCON1000,fcon1000Beijing_sub34895,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub34895.nii.gz
+769,FCON1000,fcon1000Beijing_sub34943,21.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub34943.nii.gz
+770,FCON1000,fcon1000Beijing_sub35309,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub35309.nii.gz
+771,FCON1000,fcon1000Beijing_sub35776,24.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub35776.nii.gz
+772,FCON1000,fcon1000Beijing_sub35806,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub35806.nii.gz
+773,FCON1000,fcon1000Beijing_sub36580,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub36580.nii.gz
+774,FCON1000,fcon1000Beijing_sub36942,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub36942.nii.gz
+775,FCON1000,fcon1000Beijing_sub37602,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub37602.nii.gz
+776,FCON1000,fcon1000Beijing_sub38602,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub38602.nii.gz
+777,FCON1000,fcon1000Beijing_sub39725,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub39725.nii.gz
+778,FCON1000,fcon1000Beijing_sub40037,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub40037.nii.gz
+779,FCON1000,fcon1000Beijing_sub40427,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub40427.nii.gz
+780,FCON1000,fcon1000Beijing_sub41170,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub41170.nii.gz
+781,FCON1000,fcon1000Beijing_sub41621,19.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub41621.nii.gz
+782,FCON1000,fcon1000Beijing_sub42512,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub42512.nii.gz
+783,FCON1000,fcon1000Beijing_sub42555,18.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub42555.nii.gz
+784,FCON1000,fcon1000Beijing_sub42843,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub42843.nii.gz
+785,FCON1000,fcon1000Beijing_sub43290,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub43290.nii.gz
+786,FCON1000,fcon1000Beijing_sub44573,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub44573.nii.gz
+787,FCON1000,fcon1000Beijing_sub45552,18.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub45552.nii.gz
+788,FCON1000,fcon1000Beijing_sub46058,25.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub46058.nii.gz
+789,FCON1000,fcon1000Beijing_sub46259,19.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub46259.nii.gz
+790,FCON1000,fcon1000Beijing_sub46541,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub46541.nii.gz
+791,FCON1000,fcon1000Beijing_sub48501,23.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub48501.nii.gz
+792,FCON1000,fcon1000Beijing_sub48563,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub48563.nii.gz
+793,FCON1000,fcon1000Beijing_sub48676,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub48676.nii.gz
+794,FCON1000,fcon1000Beijing_sub49782,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub49782.nii.gz
+795,FCON1000,fcon1000Beijing_sub50498,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub50498.nii.gz
+796,FCON1000,fcon1000Beijing_sub50873,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub50873.nii.gz
+797,FCON1000,fcon1000Beijing_sub50972,24.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub50972.nii.gz
+798,FCON1000,fcon1000Beijing_sub50985,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub50985.nii.gz
+799,FCON1000,fcon1000Beijing_sub51015,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub51015.nii.gz
+800,FCON1000,fcon1000Beijing_sub51586,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub51586.nii.gz
+801,FCON1000,fcon1000Beijing_sub52044,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub52044.nii.gz
+802,FCON1000,fcon1000Beijing_sub52259,21.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub52259.nii.gz
+803,FCON1000,fcon1000Beijing_sub53572,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub53572.nii.gz
+804,FCON1000,fcon1000Beijing_sub53998,20.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub53998.nii.gz
+805,FCON1000,fcon1000Beijing_sub54890,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub54890.nii.gz
+806,FCON1000,fcon1000Beijing_sub55301,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub55301.nii.gz
+807,FCON1000,fcon1000Beijing_sub55541,25.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub55541.nii.gz
+808,FCON1000,fcon1000Beijing_sub55736,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub55736.nii.gz
+809,FCON1000,fcon1000Beijing_sub55856,22.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub55856.nii.gz
+810,FCON1000,fcon1000Beijing_sub56136,22.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub56136.nii.gz
+811,FCON1000,fcon1000Beijing_sub56659,23.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub56659.nii.gz
+812,FCON1000,fcon1000Beijing_sub56703,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub56703.nii.gz
+813,FCON1000,fcon1000Beijing_sub56757,21.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub56757.nii.gz
+814,FCON1000,fcon1000Beijing_sub58029,26.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub58029.nii.gz
+815,FCON1000,fcon1000Beijing_sub58332,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub58332.nii.gz
+816,FCON1000,fcon1000Beijing_sub58614,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub58614.nii.gz
+817,FCON1000,fcon1000Beijing_sub59347,21.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub59347.nii.gz
+818,FCON1000,fcon1000Beijing_sub59448,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub59448.nii.gz
+819,FCON1000,fcon1000Beijing_sub61961,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub61961.nii.gz
+820,FCON1000,fcon1000Beijing_sub62083,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub62083.nii.gz
+821,FCON1000,fcon1000Beijing_sub62438,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub62438.nii.gz
+822,FCON1000,fcon1000Beijing_sub62843,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub62843.nii.gz
+823,FCON1000,fcon1000Beijing_sub62966,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub62966.nii.gz
+824,FCON1000,fcon1000Beijing_sub64923,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub64923.nii.gz
+825,FCON1000,fcon1000Beijing_sub65467,21.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub65467.nii.gz
+826,FCON1000,fcon1000Beijing_sub65659,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub65659.nii.gz
+827,FCON1000,fcon1000Beijing_sub66158,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub66158.nii.gz
+828,FCON1000,fcon1000Beijing_sub66528,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub66528.nii.gz
+829,FCON1000,fcon1000Beijing_sub66781,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub66781.nii.gz
+830,FCON1000,fcon1000Beijing_sub66889,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub66889.nii.gz
+831,FCON1000,fcon1000Beijing_sub67435,20.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub67435.nii.gz
+832,FCON1000,fcon1000Beijing_sub67844,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub67844.nii.gz
+833,FCON1000,fcon1000Beijing_sub68012,24.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub68012.nii.gz
+834,FCON1000,fcon1000Beijing_sub68597,19.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub68597.nii.gz
+835,FCON1000,fcon1000Beijing_sub69518,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub69518.nii.gz
+836,FCON1000,fcon1000Beijing_sub69696,18.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub69696.nii.gz
+837,FCON1000,fcon1000Beijing_sub71693,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub71693.nii.gz
+838,FCON1000,fcon1000Beijing_sub72654,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub72654.nii.gz
+839,FCON1000,fcon1000Beijing_sub72678,20.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub72678.nii.gz
+840,FCON1000,fcon1000Beijing_sub73098,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub73098.nii.gz
+841,FCON1000,fcon1000Beijing_sub73245,24.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub73245.nii.gz
+842,FCON1000,fcon1000Beijing_sub73279,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub73279.nii.gz
+843,FCON1000,fcon1000Beijing_sub73421,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub73421.nii.gz
+844,FCON1000,fcon1000Beijing_sub74386,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub74386.nii.gz
+845,FCON1000,fcon1000Beijing_sub74587,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub74587.nii.gz
+846,FCON1000,fcon1000Beijing_sub75878,24.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub75878.nii.gz
+847,FCON1000,fcon1000Beijing_sub76377,24.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub76377.nii.gz
+848,FCON1000,fcon1000Beijing_sub77440,24.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub77440.nii.gz
+849,FCON1000,fcon1000Beijing_sub80163,24.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub80163.nii.gz
+850,FCON1000,fcon1000Beijing_sub80551,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub80551.nii.gz
+851,FCON1000,fcon1000Beijing_sub80569,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub80569.nii.gz
+852,FCON1000,fcon1000Beijing_sub80927,19.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub80927.nii.gz
+853,FCON1000,fcon1000Beijing_sub81062,22.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub81062.nii.gz
+854,FCON1000,fcon1000Beijing_sub81074,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub81074.nii.gz
+855,FCON1000,fcon1000Beijing_sub82352,24.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub82352.nii.gz
+856,FCON1000,fcon1000Beijing_sub82426,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub82426.nii.gz
+857,FCON1000,fcon1000Beijing_sub82714,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub82714.nii.gz
+858,FCON1000,fcon1000Beijing_sub82826,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub82826.nii.gz
+859,FCON1000,fcon1000Beijing_sub82980,20.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub82980.nii.gz
+860,FCON1000,fcon1000Beijing_sub83430,18.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub83430.nii.gz
+861,FCON1000,fcon1000Beijing_sub83624,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub83624.nii.gz
+862,FCON1000,fcon1000Beijing_sub83728,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub83728.nii.gz
+863,FCON1000,fcon1000Beijing_sub85030,22.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub85030.nii.gz
+864,FCON1000,fcon1000Beijing_sub85543,20.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub85543.nii.gz
+865,FCON1000,fcon1000Beijing_sub85818,23.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub85818.nii.gz
+866,FCON1000,fcon1000Beijing_sub86114,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub86114.nii.gz
+867,FCON1000,fcon1000Beijing_sub87089,24.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub87089.nii.gz
+868,FCON1000,fcon1000Beijing_sub87776,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub87776.nii.gz
+869,FCON1000,fcon1000Beijing_sub88306,19.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub88306.nii.gz
+870,FCON1000,fcon1000Beijing_sub88947,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub88947.nii.gz
+871,FCON1000,fcon1000Beijing_sub89088,19.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub89088.nii.gz
+872,FCON1000,fcon1000Beijing_sub89238,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub89238.nii.gz
+873,FCON1000,fcon1000Beijing_sub89592,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub89592.nii.gz
+874,FCON1000,fcon1000Beijing_sub89742,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub89742.nii.gz
+875,FCON1000,fcon1000Beijing_sub89941,18.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub89941.nii.gz
+876,FCON1000,fcon1000Beijing_sub91145,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub91145.nii.gz
+877,FCON1000,fcon1000Beijing_sub91399,21.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub91399.nii.gz
+878,FCON1000,fcon1000Beijing_sub91952,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub91952.nii.gz
+879,FCON1000,fcon1000Beijing_sub92430,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub92430.nii.gz
+880,FCON1000,fcon1000Beijing_sub92490,20.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub92490.nii.gz
+881,FCON1000,fcon1000Beijing_sub92544,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub92544.nii.gz
+882,FCON1000,fcon1000Beijing_sub92602,26.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub92602.nii.gz
+883,FCON1000,fcon1000Beijing_sub92799,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub92799.nii.gz
+884,FCON1000,fcon1000Beijing_sub92859,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub92859.nii.gz
+885,FCON1000,fcon1000Beijing_sub93689,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub93689.nii.gz
+886,FCON1000,fcon1000Beijing_sub93856,20.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub93856.nii.gz
+887,FCON1000,fcon1000Beijing_sub94536,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub94536.nii.gz
+888,FCON1000,fcon1000Beijing_sub95575,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub95575.nii.gz
+889,FCON1000,fcon1000Beijing_sub95755,26.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub95755.nii.gz
+890,FCON1000,fcon1000Beijing_sub96163,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub96163.nii.gz
+891,FCON1000,fcon1000Beijing_sub97442,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub97442.nii.gz
+892,FCON1000,fcon1000Beijing_sub98353,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub98353.nii.gz
+893,FCON1000,fcon1000Beijing_sub98617,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Beijing_sub98617.nii.gz
+894,FCON1000,fcon1000Berlin_sub06204,34.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub06204.nii.gz
+895,FCON1000,fcon1000Berlin_sub06716,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub06716.nii.gz
+896,FCON1000,fcon1000Berlin_sub12855,33.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub12855.nii.gz
+897,FCON1000,fcon1000Berlin_sub18913,29.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub18913.nii.gz
+898,FCON1000,fcon1000Berlin_sub23506,27.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub23506.nii.gz
+899,FCON1000,fcon1000Berlin_sub27536,25.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub27536.nii.gz
+900,FCON1000,fcon1000Berlin_sub27711,26.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub27711.nii.gz
+901,FCON1000,fcon1000Berlin_sub27797,31.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub27797.nii.gz
+902,FCON1000,fcon1000Berlin_sub28092,26.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub28092.nii.gz
+903,FCON1000,fcon1000Berlin_sub33248,28.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub33248.nii.gz
+904,FCON1000,fcon1000Berlin_sub38279,29.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub38279.nii.gz
+905,FCON1000,fcon1000Berlin_sub40143,34.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub40143.nii.gz
+906,FCON1000,fcon1000Berlin_sub47066,26.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub47066.nii.gz
+907,FCON1000,fcon1000Berlin_sub47791,31.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub47791.nii.gz
+908,FCON1000,fcon1000Berlin_sub49134,44.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub49134.nii.gz
+909,FCON1000,fcon1000Berlin_sub54976,37.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub54976.nii.gz
+910,FCON1000,fcon1000Berlin_sub57028,37.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub57028.nii.gz
+911,FCON1000,fcon1000Berlin_sub67166,32.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub67166.nii.gz
+912,FCON1000,fcon1000Berlin_sub75506,28.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub75506.nii.gz
+913,FCON1000,fcon1000Berlin_sub77281,24.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub77281.nii.gz
+914,FCON1000,fcon1000Berlin_sub85681,26.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub85681.nii.gz
+915,FCON1000,fcon1000Berlin_sub86111,24.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub86111.nii.gz
+916,FCON1000,fcon1000Berlin_sub91116,39.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub91116.nii.gz
+917,FCON1000,fcon1000Berlin_sub91966,27.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub91966.nii.gz
+918,FCON1000,fcon1000Berlin_sub95068,26.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub95068.nii.gz
+919,FCON1000,fcon1000Berlin_sub97162,28.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Berlin_sub97162.nii.gz
+920,FCON1000,fcon1000Cambridge_sub00156,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub00156.nii.gz
+921,FCON1000,fcon1000Cambridge_sub00294,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub00294.nii.gz
+922,FCON1000,fcon1000Cambridge_sub01361,19.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub01361.nii.gz
+923,FCON1000,fcon1000Cambridge_sub02591,22.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub02591.nii.gz
+924,FCON1000,fcon1000Cambridge_sub02953,23.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub02953.nii.gz
+925,FCON1000,fcon1000Cambridge_sub04187,18.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub04187.nii.gz
+926,FCON1000,fcon1000Cambridge_sub04270,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub04270.nii.gz
+927,FCON1000,fcon1000Cambridge_sub04491,25.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub04491.nii.gz
+928,FCON1000,fcon1000Cambridge_sub04665,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub04665.nii.gz
+929,FCON1000,fcon1000Cambridge_sub05306,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub05306.nii.gz
+930,FCON1000,fcon1000Cambridge_sub05453,20.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub05453.nii.gz
+931,FCON1000,fcon1000Cambridge_sub06037,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub06037.nii.gz
+932,FCON1000,fcon1000Cambridge_sub06272,22.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub06272.nii.gz
+933,FCON1000,fcon1000Cambridge_sub06987,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub06987.nii.gz
+934,FCON1000,fcon1000Cambridge_sub07413,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub07413.nii.gz
+935,FCON1000,fcon1000Cambridge_sub07798,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub07798.nii.gz
+936,FCON1000,fcon1000Cambridge_sub07902,23.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub07902.nii.gz
+937,FCON1000,fcon1000Cambridge_sub08204,26.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub08204.nii.gz
+938,FCON1000,fcon1000Cambridge_sub08588,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub08588.nii.gz
+939,FCON1000,fcon1000Cambridge_sub08723,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub08723.nii.gz
+940,FCON1000,fcon1000Cambridge_sub08947,19.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub08947.nii.gz
+941,FCON1000,fcon1000Cambridge_sub09015,25.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub09015.nii.gz
+942,FCON1000,fcon1000Cambridge_sub09397,30.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub09397.nii.gz
+943,FCON1000,fcon1000Cambridge_sub09633,30.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub09633.nii.gz
+944,FCON1000,fcon1000Cambridge_sub10268,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub10268.nii.gz
+945,FCON1000,fcon1000Cambridge_sub10619,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub10619.nii.gz
+946,FCON1000,fcon1000Cambridge_sub11388,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub11388.nii.gz
+947,FCON1000,fcon1000Cambridge_sub12346,18.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub12346.nii.gz
+948,FCON1000,fcon1000Cambridge_sub13093,19.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub13093.nii.gz
+949,FCON1000,fcon1000Cambridge_sub13187,23.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub13187.nii.gz
+950,FCON1000,fcon1000Cambridge_sub13216,18.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub13216.nii.gz
+951,FCON1000,fcon1000Cambridge_sub13902,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub13902.nii.gz
+952,FCON1000,fcon1000Cambridge_sub14183,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub14183.nii.gz
+953,FCON1000,fcon1000Cambridge_sub14194,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub14194.nii.gz
+954,FCON1000,fcon1000Cambridge_sub14278,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub14278.nii.gz
+955,FCON1000,fcon1000Cambridge_sub15172,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub15172.nii.gz
+956,FCON1000,fcon1000Cambridge_sub15258,23.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub15258.nii.gz
+957,FCON1000,fcon1000Cambridge_sub15432,23.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub15432.nii.gz
+958,FCON1000,fcon1000Cambridge_sub15905,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub15905.nii.gz
+959,FCON1000,fcon1000Cambridge_sub16122,18.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub16122.nii.gz
+960,FCON1000,fcon1000Cambridge_sub16390,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub16390.nii.gz
+961,FCON1000,fcon1000Cambridge_sub16846,29.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub16846.nii.gz
+962,FCON1000,fcon1000Cambridge_sub17584,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub17584.nii.gz
+963,FCON1000,fcon1000Cambridge_sub17737,26.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub17737.nii.gz
+964,FCON1000,fcon1000Cambridge_sub17772,24.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub17772.nii.gz
+965,FCON1000,fcon1000Cambridge_sub18295,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub18295.nii.gz
+966,FCON1000,fcon1000Cambridge_sub18449,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub18449.nii.gz
+967,FCON1000,fcon1000Cambridge_sub19717,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub19717.nii.gz
+968,FCON1000,fcon1000Cambridge_sub20389,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub20389.nii.gz
+969,FCON1000,fcon1000Cambridge_sub20543,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub20543.nii.gz
+970,FCON1000,fcon1000Cambridge_sub20563,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub20563.nii.gz
+971,FCON1000,fcon1000Cambridge_sub21755,21.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub21755.nii.gz
+972,FCON1000,fcon1000Cambridge_sub23780,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub23780.nii.gz
+973,FCON1000,fcon1000Cambridge_sub23869,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub23869.nii.gz
+974,FCON1000,fcon1000Cambridge_sub24670,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub24670.nii.gz
+975,FCON1000,fcon1000Cambridge_sub24757,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub24757.nii.gz
+976,FCON1000,fcon1000Cambridge_sub25044,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub25044.nii.gz
+977,FCON1000,fcon1000Cambridge_sub25058,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub25058.nii.gz
+978,FCON1000,fcon1000Cambridge_sub25099,26.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub25099.nii.gz
+979,FCON1000,fcon1000Cambridge_sub26348,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub26348.nii.gz
+980,FCON1000,fcon1000Cambridge_sub27065,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub27065.nii.gz
+981,FCON1000,fcon1000Cambridge_sub27230,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub27230.nii.gz
+982,FCON1000,fcon1000Cambridge_sub27613,18.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub27613.nii.gz
+983,FCON1000,fcon1000Cambridge_sub27796,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub27796.nii.gz
+984,FCON1000,fcon1000Cambridge_sub29044,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub29044.nii.gz
+985,FCON1000,fcon1000Cambridge_sub29425,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub29425.nii.gz
+986,FCON1000,fcon1000Cambridge_sub29800,24.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub29800.nii.gz
+987,FCON1000,fcon1000Cambridge_sub31522,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub31522.nii.gz
+988,FCON1000,fcon1000Cambridge_sub34586,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub34586.nii.gz
+989,FCON1000,fcon1000Cambridge_sub34741,27.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub34741.nii.gz
+990,FCON1000,fcon1000Cambridge_sub35430,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub35430.nii.gz
+991,FCON1000,fcon1000Cambridge_sub35512,19.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub35512.nii.gz
+992,FCON1000,fcon1000Cambridge_sub37374,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub37374.nii.gz
+993,FCON1000,fcon1000Cambridge_sub39053,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub39053.nii.gz
+994,FCON1000,fcon1000Cambridge_sub39065,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub39065.nii.gz
+995,FCON1000,fcon1000Cambridge_sub39142,25.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub39142.nii.gz
+996,FCON1000,fcon1000Cambridge_sub39737,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub39737.nii.gz
+997,FCON1000,fcon1000Cambridge_sub40635,18.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub40635.nii.gz
+998,FCON1000,fcon1000Cambridge_sub41567,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub41567.nii.gz
+999,FCON1000,fcon1000Cambridge_sub41773,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub41773.nii.gz
+1000,FCON1000,fcon1000Cambridge_sub41814,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub41814.nii.gz
+1001,FCON1000,fcon1000Cambridge_sub42146,24.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub42146.nii.gz
+1002,FCON1000,fcon1000Cambridge_sub42253,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub42253.nii.gz
+1003,FCON1000,fcon1000Cambridge_sub43304,23.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub43304.nii.gz
+1004,FCON1000,fcon1000Cambridge_sub43358,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub43358.nii.gz
+1005,FCON1000,fcon1000Cambridge_sub45344,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub45344.nii.gz
+1006,FCON1000,fcon1000Cambridge_sub45354,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub45354.nii.gz
+1007,FCON1000,fcon1000Cambridge_sub45604,23.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub45604.nii.gz
+1008,FCON1000,fcon1000Cambridge_sub47162,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub47162.nii.gz
+1009,FCON1000,fcon1000Cambridge_sub47231,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub47231.nii.gz
+1010,FCON1000,fcon1000Cambridge_sub47278,20.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub47278.nii.gz
+1011,FCON1000,fcon1000Cambridge_sub47498,19.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub47498.nii.gz
+1012,FCON1000,fcon1000Cambridge_sub49259,19.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub49259.nii.gz
+1013,FCON1000,fcon1000Cambridge_sub49998,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub49998.nii.gz
+1014,FCON1000,fcon1000Cambridge_sub50272,18.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub50272.nii.gz
+1015,FCON1000,fcon1000Cambridge_sub50454,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub50454.nii.gz
+1016,FCON1000,fcon1000Cambridge_sub50953,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub50953.nii.gz
+1017,FCON1000,fcon1000Cambridge_sub51050,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub51050.nii.gz
+1018,FCON1000,fcon1000Cambridge_sub51172,18.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub51172.nii.gz
+1019,FCON1000,fcon1000Cambridge_sub51512,23.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub51512.nii.gz
+1020,FCON1000,fcon1000Cambridge_sub51671,22.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub51671.nii.gz
+1021,FCON1000,fcon1000Cambridge_sub52036,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub52036.nii.gz
+1022,FCON1000,fcon1000Cambridge_sub52300,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub52300.nii.gz
+1023,FCON1000,fcon1000Cambridge_sub52442,25.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub52442.nii.gz
+1024,FCON1000,fcon1000Cambridge_sub53059,22.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub53059.nii.gz
+1025,FCON1000,fcon1000Cambridge_sub53107,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub53107.nii.gz
+1026,FCON1000,fcon1000Cambridge_sub53193,20.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub53193.nii.gz
+1027,FCON1000,fcon1000Cambridge_sub53296,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub53296.nii.gz
+1028,FCON1000,fcon1000Cambridge_sub53615,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub53615.nii.gz
+1029,FCON1000,fcon1000Cambridge_sub54846,20.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub54846.nii.gz
+1030,FCON1000,fcon1000Cambridge_sub55114,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub55114.nii.gz
+1031,FCON1000,fcon1000Cambridge_sub55660,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub55660.nii.gz
+1032,FCON1000,fcon1000Cambridge_sub55874,19.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub55874.nii.gz
+1033,FCON1000,fcon1000Cambridge_sub57221,21.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub57221.nii.gz
+1034,FCON1000,fcon1000Cambridge_sub58360,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub58360.nii.gz
+1035,FCON1000,fcon1000Cambridge_sub58470,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub58470.nii.gz
+1036,FCON1000,fcon1000Cambridge_sub58682,22.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub58682.nii.gz
+1037,FCON1000,fcon1000Cambridge_sub58874,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub58874.nii.gz
+1038,FCON1000,fcon1000Cambridge_sub59434,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub59434.nii.gz
+1039,FCON1000,fcon1000Cambridge_sub59729,23.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub59729.nii.gz
+1040,FCON1000,fcon1000Cambridge_sub60578,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub60578.nii.gz
+1041,FCON1000,fcon1000Cambridge_sub60797,19.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub60797.nii.gz
+1042,FCON1000,fcon1000Cambridge_sub61185,26.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub61185.nii.gz
+1043,FCON1000,fcon1000Cambridge_sub61209,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub61209.nii.gz
+1044,FCON1000,fcon1000Cambridge_sub61436,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub61436.nii.gz
+1045,FCON1000,fcon1000Cambridge_sub61753,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub61753.nii.gz
+1046,FCON1000,fcon1000Cambridge_sub62383,22.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub62383.nii.gz
+1047,FCON1000,fcon1000Cambridge_sub62424,20.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub62424.nii.gz
+1048,FCON1000,fcon1000Cambridge_sub62732,18.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub62732.nii.gz
+1049,FCON1000,fcon1000Cambridge_sub62908,26.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub62908.nii.gz
+1050,FCON1000,fcon1000Cambridge_sub63412,19.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub63412.nii.gz
+1051,FCON1000,fcon1000Cambridge_sub63661,23.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub63661.nii.gz
+1052,FCON1000,fcon1000Cambridge_sub64308,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub64308.nii.gz
+1053,FCON1000,fcon1000Cambridge_sub64985,18.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub64985.nii.gz
+1054,FCON1000,fcon1000Cambridge_sub65373,18.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub65373.nii.gz
+1055,FCON1000,fcon1000Cambridge_sub65682,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub65682.nii.gz
+1056,FCON1000,fcon1000Cambridge_sub66351,18.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub66351.nii.gz
+1057,FCON1000,fcon1000Cambridge_sub67117,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub67117.nii.gz
+1058,FCON1000,fcon1000Cambridge_sub68101,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub68101.nii.gz
+1059,FCON1000,fcon1000Cambridge_sub68425,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub68425.nii.gz
+1060,FCON1000,fcon1000Cambridge_sub69287,18.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub69287.nii.gz
+1061,FCON1000,fcon1000Cambridge_sub69315,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub69315.nii.gz
+1062,FCON1000,fcon1000Cambridge_sub69397,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub69397.nii.gz
+1063,FCON1000,fcon1000Cambridge_sub71849,21.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub71849.nii.gz
+1064,FCON1000,fcon1000Cambridge_sub72068,18.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub72068.nii.gz
+1065,FCON1000,fcon1000Cambridge_sub73317,19.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub73317.nii.gz
+1066,FCON1000,fcon1000Cambridge_sub73399,18.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub73399.nii.gz
+1067,FCON1000,fcon1000Cambridge_sub73477,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub73477.nii.gz
+1068,FCON1000,fcon1000Cambridge_sub76631,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub76631.nii.gz
+1069,FCON1000,fcon1000Cambridge_sub76745,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub76745.nii.gz
+1070,FCON1000,fcon1000Cambridge_sub77337,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub77337.nii.gz
+1071,FCON1000,fcon1000Cambridge_sub77435,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub77435.nii.gz
+1072,FCON1000,fcon1000Cambridge_sub77598,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub77598.nii.gz
+1073,FCON1000,fcon1000Cambridge_sub77989,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub77989.nii.gz
+1074,FCON1000,fcon1000Cambridge_sub78547,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub78547.nii.gz
+1075,FCON1000,fcon1000Cambridge_sub78552,19.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub78552.nii.gz
+1076,FCON1000,fcon1000Cambridge_sub78614,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub78614.nii.gz
+1077,FCON1000,fcon1000Cambridge_sub80557,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub80557.nii.gz
+1078,FCON1000,fcon1000Cambridge_sub81289,19.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub81289.nii.gz
+1079,FCON1000,fcon1000Cambridge_sub81524,24.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub81524.nii.gz
+1080,FCON1000,fcon1000Cambridge_sub81562,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub81562.nii.gz
+1081,FCON1000,fcon1000Cambridge_sub82113,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub82113.nii.gz
+1082,FCON1000,fcon1000Cambridge_sub82213,20.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub82213.nii.gz
+1083,FCON1000,fcon1000Cambridge_sub82435,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub82435.nii.gz
+1084,FCON1000,fcon1000Cambridge_sub83409,18.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub83409.nii.gz
+1085,FCON1000,fcon1000Cambridge_sub83683,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub83683.nii.gz
+1086,FCON1000,fcon1000Cambridge_sub84064,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub84064.nii.gz
+1087,FCON1000,fcon1000Cambridge_sub84256,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub84256.nii.gz
+1088,FCON1000,fcon1000Cambridge_sub84504,18.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub84504.nii.gz
+1089,FCON1000,fcon1000Cambridge_sub84845,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub84845.nii.gz
+1090,FCON1000,fcon1000Cambridge_sub86115,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub86115.nii.gz
+1091,FCON1000,fcon1000Cambridge_sub86637,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub86637.nii.gz
+1092,FCON1000,fcon1000Cambridge_sub87846,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub87846.nii.gz
+1093,FCON1000,fcon1000Cambridge_sub88445,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub88445.nii.gz
+1094,FCON1000,fcon1000Cambridge_sub88464,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub88464.nii.gz
+1095,FCON1000,fcon1000Cambridge_sub88853,29.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub88853.nii.gz
+1096,FCON1000,fcon1000Cambridge_sub89107,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub89107.nii.gz
+1097,FCON1000,fcon1000Cambridge_sub89435,18.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub89435.nii.gz
+1098,FCON1000,fcon1000Cambridge_sub89894,18.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub89894.nii.gz
+1099,FCON1000,fcon1000Cambridge_sub90059,19.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub90059.nii.gz
+1100,FCON1000,fcon1000Cambridge_sub90674,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub90674.nii.gz
+1101,FCON1000,fcon1000Cambridge_sub90681,22.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub90681.nii.gz
+1102,FCON1000,fcon1000Cambridge_sub90699,26.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub90699.nii.gz
+1103,FCON1000,fcon1000Cambridge_sub92288,23.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub92288.nii.gz
+1104,FCON1000,fcon1000Cambridge_sub92393,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub92393.nii.gz
+1105,FCON1000,fcon1000Cambridge_sub92440,24.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub92440.nii.gz
+1106,FCON1000,fcon1000Cambridge_sub93269,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub93269.nii.gz
+1107,FCON1000,fcon1000Cambridge_sub93488,20.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub93488.nii.gz
+1108,FCON1000,fcon1000Cambridge_sub93609,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub93609.nii.gz
+1109,FCON1000,fcon1000Cambridge_sub94304,22.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub94304.nii.gz
+1110,FCON1000,fcon1000Cambridge_sub95187,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub95187.nii.gz
+1111,FCON1000,fcon1000Cambridge_sub95644,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub95644.nii.gz
+1112,FCON1000,fcon1000Cambridge_sub95959,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub95959.nii.gz
+1113,FCON1000,fcon1000Cambridge_sub98528,19.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub98528.nii.gz
+1114,FCON1000,fcon1000Cambridge_sub98624,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub98624.nii.gz
+1115,FCON1000,fcon1000Cambridge_sub99085,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub99085.nii.gz
+1116,FCON1000,fcon1000Cambridge_sub99330,18.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub99330.nii.gz
+1117,FCON1000,fcon1000Cambridge_sub99462,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cambridge_sub99462.nii.gz
+1118,FCON1000,fcon1000Cleveland_sub02480,56.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub02480.nii.gz
+1119,FCON1000,fcon1000Cleveland_sub07835,43.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub07835.nii.gz
+1120,FCON1000,fcon1000Cleveland_sub12330,54.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub12330.nii.gz
+1121,FCON1000,fcon1000Cleveland_sub13495,45.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub13495.nii.gz
+1122,FCON1000,fcon1000Cleveland_sub17946,29.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub17946.nii.gz
+1123,FCON1000,fcon1000Cleveland_sub18011,55.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub18011.nii.gz
+1124,FCON1000,fcon1000Cleveland_sub18566,33.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub18566.nii.gz
+1125,FCON1000,fcon1000Cleveland_sub19005,29.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub19005.nii.gz
+1126,FCON1000,fcon1000Cleveland_sub20003,39.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub20003.nii.gz
+1127,FCON1000,fcon1000Cleveland_sub22736,28.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub22736.nii.gz
+1128,FCON1000,fcon1000Cleveland_sub22935,39.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub22935.nii.gz
+1129,FCON1000,fcon1000Cleveland_sub26557,51.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub26557.nii.gz
+1130,FCON1000,fcon1000Cleveland_sub28596,24.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub28596.nii.gz
+1131,FCON1000,fcon1000Cleveland_sub34189,55.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub34189.nii.gz
+1132,FCON1000,fcon1000Cleveland_sub46075,33.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub46075.nii.gz
+1133,FCON1000,fcon1000Cleveland_sub46739,60.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub46739.nii.gz
+1134,FCON1000,fcon1000Cleveland_sub47482,28.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub47482.nii.gz
+1135,FCON1000,fcon1000Cleveland_sub50092,56.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub50092.nii.gz
+1136,FCON1000,fcon1000Cleveland_sub58811,36.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub58811.nii.gz
+1137,FCON1000,fcon1000Cleveland_sub61868,48.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub61868.nii.gz
+1138,FCON1000,fcon1000Cleveland_sub64706,57.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub64706.nii.gz
+1139,FCON1000,fcon1000Cleveland_sub65858,53.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub65858.nii.gz
+1140,FCON1000,fcon1000Cleveland_sub67936,57.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub67936.nii.gz
+1141,FCON1000,fcon1000Cleveland_sub75398,43.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub75398.nii.gz
+1142,FCON1000,fcon1000Cleveland_sub76139,41.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub76139.nii.gz
+1143,FCON1000,fcon1000Cleveland_sub80263,44.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub80263.nii.gz
+1144,FCON1000,fcon1000Cleveland_sub82518,30.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub82518.nii.gz
+1145,FCON1000,fcon1000Cleveland_sub85091,52.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub85091.nii.gz
+1146,FCON1000,fcon1000Cleveland_sub92232,45.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub92232.nii.gz
+1147,FCON1000,fcon1000Cleveland_sub97844,30.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub97844.nii.gz
+1148,FCON1000,fcon1000Cleveland_sub99664,57.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Cleveland_sub99664.nii.gz
+1149,FCON1000,fcon1000ICBM_sub00448,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub00448.nii.gz
+1150,FCON1000,fcon1000ICBM_sub00623,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub00623.nii.gz
+1151,FCON1000,fcon1000ICBM_sub02382,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub02382.nii.gz
+1152,FCON1000,fcon1000ICBM_sub02503,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub02503.nii.gz
+1153,FCON1000,fcon1000ICBM_sub05208,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub05208.nii.gz
+1154,FCON1000,fcon1000ICBM_sub07286,,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub07286.nii.gz
+1155,FCON1000,fcon1000ICBM_sub08255,,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub08255.nii.gz
+1156,FCON1000,fcon1000ICBM_sub08806,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub08806.nii.gz
+1157,FCON1000,fcon1000ICBM_sub09539,,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub09539.nii.gz
+1158,FCON1000,fcon1000ICBM_sub10582,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub10582.nii.gz
+1159,FCON1000,fcon1000ICBM_sub13384,,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub13384.nii.gz
+1160,FCON1000,fcon1000ICBM_sub13478,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub13478.nii.gz
+1161,FCON1000,fcon1000ICBM_sub13789,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub13789.nii.gz
+1162,FCON1000,fcon1000ICBM_sub16607,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub16607.nii.gz
+1163,FCON1000,fcon1000ICBM_sub19395,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub19395.nii.gz
+1164,FCON1000,fcon1000ICBM_sub20718,,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub20718.nii.gz
+1165,FCON1000,fcon1000ICBM_sub22674,,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub22674.nii.gz
+1166,FCON1000,fcon1000ICBM_sub26183,,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub26183.nii.gz
+1167,FCON1000,fcon1000ICBM_sub26796,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub26796.nii.gz
+1168,FCON1000,fcon1000ICBM_sub28422,,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub28422.nii.gz
+1169,FCON1000,fcon1000ICBM_sub28795,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub28795.nii.gz
+1170,FCON1000,fcon1000ICBM_sub28808,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub28808.nii.gz
+1171,FCON1000,fcon1000ICBM_sub29353,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub29353.nii.gz
+1172,FCON1000,fcon1000ICBM_sub30003,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub30003.nii.gz
+1173,FCON1000,fcon1000ICBM_sub30623,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub30623.nii.gz
+1174,FCON1000,fcon1000ICBM_sub32549,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub32549.nii.gz
+1175,FCON1000,fcon1000ICBM_sub33677,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub33677.nii.gz
+1176,FCON1000,fcon1000ICBM_sub34252,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub34252.nii.gz
+1177,FCON1000,fcon1000ICBM_sub35262,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub35262.nii.gz
+1178,FCON1000,fcon1000ICBM_sub35370,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub35370.nii.gz
+1179,FCON1000,fcon1000ICBM_sub37140,,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub37140.nii.gz
+1180,FCON1000,fcon1000ICBM_sub40217,,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub40217.nii.gz
+1181,FCON1000,fcon1000ICBM_sub40482,,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub40482.nii.gz
+1182,FCON1000,fcon1000ICBM_sub41546,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub41546.nii.gz
+1183,FCON1000,fcon1000ICBM_sub41764,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub41764.nii.gz
+1184,FCON1000,fcon1000ICBM_sub44077,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub44077.nii.gz
+1185,FCON1000,fcon1000ICBM_sub44395,,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub44395.nii.gz
+1186,FCON1000,fcon1000ICBM_sub47658,,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub47658.nii.gz
+1187,FCON1000,fcon1000ICBM_sub47753,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub47753.nii.gz
+1188,FCON1000,fcon1000ICBM_sub48210,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub48210.nii.gz
+1189,FCON1000,fcon1000ICBM_sub48830,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub48830.nii.gz
+1190,FCON1000,fcon1000ICBM_sub49215,,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub49215.nii.gz
+1191,FCON1000,fcon1000ICBM_sub51677,,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub51677.nii.gz
+1192,FCON1000,fcon1000ICBM_sub53282,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub53282.nii.gz
+1193,FCON1000,fcon1000ICBM_sub53801,,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub53801.nii.gz
+1194,FCON1000,fcon1000ICBM_sub54887,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub54887.nii.gz
+1195,FCON1000,fcon1000ICBM_sub55114,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub55114.nii.gz
+1196,FCON1000,fcon1000ICBM_sub55656,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub55656.nii.gz
+1197,FCON1000,fcon1000ICBM_sub57738,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub57738.nii.gz
+1198,FCON1000,fcon1000ICBM_sub59589,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub59589.nii.gz
+1199,FCON1000,fcon1000ICBM_sub59739,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub59739.nii.gz
+1200,FCON1000,fcon1000ICBM_sub59914,,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub59914.nii.gz
+1201,FCON1000,fcon1000ICBM_sub62937,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub62937.nii.gz
+1202,FCON1000,fcon1000ICBM_sub63280,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub63280.nii.gz
+1203,FCON1000,fcon1000ICBM_sub65921,,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub65921.nii.gz
+1204,FCON1000,fcon1000ICBM_sub66085,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub66085.nii.gz
+1205,FCON1000,fcon1000ICBM_sub66794,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub66794.nii.gz
+1206,FCON1000,fcon1000ICBM_sub68850,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub68850.nii.gz
+1207,FCON1000,fcon1000ICBM_sub70595,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub70595.nii.gz
+1208,FCON1000,fcon1000ICBM_sub71932,,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub71932.nii.gz
+1209,FCON1000,fcon1000ICBM_sub72135,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub72135.nii.gz
+1210,FCON1000,fcon1000ICBM_sub73490,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub73490.nii.gz
+1211,FCON1000,fcon1000ICBM_sub76325,,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub76325.nii.gz
+1212,FCON1000,fcon1000ICBM_sub76678,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub76678.nii.gz
+1213,FCON1000,fcon1000ICBM_sub77431,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub77431.nii.gz
+1214,FCON1000,fcon1000ICBM_sub78297,,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub78297.nii.gz
+1215,FCON1000,fcon1000ICBM_sub82221,,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub82221.nii.gz
+1216,FCON1000,fcon1000ICBM_sub82228,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub82228.nii.gz
+1217,FCON1000,fcon1000ICBM_sub82754,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub82754.nii.gz
+1218,FCON1000,fcon1000ICBM_sub85442,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub85442.nii.gz
+1219,FCON1000,fcon1000ICBM_sub86203,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub86203.nii.gz
+1220,FCON1000,fcon1000ICBM_sub86516,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub86516.nii.gz
+1221,FCON1000,fcon1000ICBM_sub86665,,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub86665.nii.gz
+1222,FCON1000,fcon1000ICBM_sub87217,,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub87217.nii.gz
+1223,FCON1000,fcon1000ICBM_sub89049,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub89049.nii.gz
+1224,FCON1000,fcon1000ICBM_sub92028,,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub92028.nii.gz
+1225,FCON1000,fcon1000ICBM_sub93262,,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub93262.nii.gz
+1226,FCON1000,fcon1000ICBM_sub93975,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub93975.nii.gz
+1227,FCON1000,fcon1000ICBM_sub94103,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub94103.nii.gz
+1228,FCON1000,fcon1000ICBM_sub94169,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub94169.nii.gz
+1229,FCON1000,fcon1000ICBM_sub94945,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub94945.nii.gz
+1230,FCON1000,fcon1000ICBM_sub95400,,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub95400.nii.gz
+1231,FCON1000,fcon1000ICBM_sub95971,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub95971.nii.gz
+1232,FCON1000,fcon1000ICBM_sub97008,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub97008.nii.gz
+1233,FCON1000,fcon1000ICBM_sub98317,,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub98317.nii.gz
+1234,FCON1000,fcon1000ICBM_sub98802,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ICBM_sub98802.nii.gz
+1235,FCON1000,fcon1000Leiden_sub01553,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub01553.nii.gz
+1236,FCON1000,fcon1000Leiden_sub01787,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub01787.nii.gz
+1237,FCON1000,fcon1000Leiden_sub04484,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub04484.nii.gz
+1238,FCON1000,fcon1000Leiden_sub08518,21.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub08518.nii.gz
+1239,FCON1000,fcon1000Leiden_sub09796,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub09796.nii.gz
+1240,FCON1000,fcon1000Leiden_sub10481,24.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub10481.nii.gz
+1241,FCON1000,fcon1000Leiden_sub12255,25.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub12255.nii.gz
+1242,FCON1000,fcon1000Leiden_sub13537,21.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub13537.nii.gz
+1243,FCON1000,fcon1000Leiden_sub18456,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub18456.nii.gz
+1244,FCON1000,fcon1000Leiden_sub19281,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub19281.nii.gz
+1245,FCON1000,fcon1000Leiden_sub28473,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub28473.nii.gz
+1246,FCON1000,fcon1000Leiden_sub30943,18.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub30943.nii.gz
+1247,FCON1000,fcon1000Leiden_sub36743,25.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub36743.nii.gz
+1248,FCON1000,fcon1000Leiden_sub38454,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub38454.nii.gz
+1249,FCON1000,fcon1000Leiden_sub39335,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub39335.nii.gz
+1250,FCON1000,fcon1000Leiden_sub40907,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub40907.nii.gz
+1251,FCON1000,fcon1000Leiden_sub52853,27.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub52853.nii.gz
+1252,FCON1000,fcon1000Leiden_sub52922,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub52922.nii.gz
+1253,FCON1000,fcon1000Leiden_sub56299,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub56299.nii.gz
+1254,FCON1000,fcon1000Leiden_sub57187,26.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub57187.nii.gz
+1255,FCON1000,fcon1000Leiden_sub58194,27.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub58194.nii.gz
+1256,FCON1000,fcon1000Leiden_sub64642,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub64642.nii.gz
+1257,FCON1000,fcon1000Leiden_sub66131,19.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub66131.nii.gz
+1258,FCON1000,fcon1000Leiden_sub68050,25.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub68050.nii.gz
+1259,FCON1000,fcon1000Leiden_sub72247,20.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub72247.nii.gz
+1260,FCON1000,fcon1000Leiden_sub86034,21.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub86034.nii.gz
+1261,FCON1000,fcon1000Leiden_sub87320,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub87320.nii.gz
+1262,FCON1000,fcon1000Leiden_sub92061,28.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub92061.nii.gz
+1263,FCON1000,fcon1000Leiden_sub93194,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub93194.nii.gz
+1264,FCON1000,fcon1000Leiden_sub97690,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub97690.nii.gz
+1265,FCON1000,fcon1000Leiden_sub99856,24.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Leiden_sub99856.nii.gz
+1266,FCON1000,fcon1000Milwaukee_sub00917,54.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub00917.nii.gz
+1267,FCON1000,fcon1000Milwaukee_sub09931,48.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub09931.nii.gz
+1268,FCON1000,fcon1000Milwaukee_sub14055,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub14055.nii.gz
+1269,FCON1000,fcon1000Milwaukee_sub14692,53.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub14692.nii.gz
+1270,FCON1000,fcon1000Milwaukee_sub16666,47.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub16666.nii.gz
+1271,FCON1000,fcon1000Milwaukee_sub17004,57.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub17004.nii.gz
+1272,FCON1000,fcon1000Milwaukee_sub17987,47.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub17987.nii.gz
+1273,FCON1000,fcon1000Milwaukee_sub18955,53.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub18955.nii.gz
+1274,FCON1000,fcon1000Milwaukee_sub21350,61.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub21350.nii.gz
+1275,FCON1000,fcon1000Milwaukee_sub23607,48.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub23607.nii.gz
+1276,FCON1000,fcon1000Milwaukee_sub24237,52.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub24237.nii.gz
+1277,FCON1000,fcon1000Milwaukee_sub27910,,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub27910.nii.gz
+1278,FCON1000,fcon1000Milwaukee_sub28782,64.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub28782.nii.gz
+1279,FCON1000,fcon1000Milwaukee_sub28859,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub28859.nii.gz
+1280,FCON1000,fcon1000Milwaukee_sub30157,54.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub30157.nii.gz
+1281,FCON1000,fcon1000Milwaukee_sub34838,,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub34838.nii.gz
+1282,FCON1000,fcon1000Milwaukee_sub36386,57.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub36386.nii.gz
+1283,FCON1000,fcon1000Milwaukee_sub39259,53.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub39259.nii.gz
+1284,FCON1000,fcon1000Milwaukee_sub39974,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub39974.nii.gz
+1285,FCON1000,fcon1000Milwaukee_sub44129,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub44129.nii.gz
+1286,FCON1000,fcon1000Milwaukee_sub44912,49.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub44912.nii.gz
+1287,FCON1000,fcon1000Milwaukee_sub45019,44.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub45019.nii.gz
+1288,FCON1000,fcon1000Milwaukee_sub45852,63.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub45852.nii.gz
+1289,FCON1000,fcon1000Milwaukee_sub46312,51.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub46312.nii.gz
+1290,FCON1000,fcon1000Milwaukee_sub47640,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub47640.nii.gz
+1291,FCON1000,fcon1000Milwaukee_sub49975,46.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub49975.nii.gz
+1292,FCON1000,fcon1000Milwaukee_sub50276,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub50276.nii.gz
+1293,FCON1000,fcon1000Milwaukee_sub50771,53.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub50771.nii.gz
+1294,FCON1000,fcon1000Milwaukee_sub51182,57.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub51182.nii.gz
+1295,FCON1000,fcon1000Milwaukee_sub53971,47.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub53971.nii.gz
+1296,FCON1000,fcon1000Milwaukee_sub55176,63.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub55176.nii.gz
+1297,FCON1000,fcon1000Milwaukee_sub55712,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub55712.nii.gz
+1298,FCON1000,fcon1000Milwaukee_sub56084,53.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub56084.nii.gz
+1299,FCON1000,fcon1000Milwaukee_sub56108,51.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub56108.nii.gz
+1300,FCON1000,fcon1000Milwaukee_sub56333,53.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub56333.nii.gz
+1301,FCON1000,fcon1000Milwaukee_sub56582,57.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub56582.nii.gz
+1302,FCON1000,fcon1000Milwaukee_sub58677,46.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub58677.nii.gz
+1303,FCON1000,fcon1000Milwaukee_sub58967,61.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub58967.nii.gz
+1304,FCON1000,fcon1000Milwaukee_sub59359,51.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub59359.nii.gz
+1305,FCON1000,fcon1000Milwaukee_sub61779,56.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub61779.nii.gz
+1306,FCON1000,fcon1000Milwaukee_sub63196,49.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub63196.nii.gz
+1307,FCON1000,fcon1000Milwaukee_sub64463,52.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub64463.nii.gz
+1308,FCON1000,fcon1000Milwaukee_sub67948,48.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub67948.nii.gz
+1309,FCON1000,fcon1000Milwaukee_sub68961,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub68961.nii.gz
+1310,FCON1000,fcon1000Milwaukee_sub73547,55.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub73547.nii.gz
+1311,FCON1000,fcon1000Milwaukee_sub75919,50.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub75919.nii.gz
+1312,FCON1000,fcon1000Milwaukee_sub76042,58.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub76042.nii.gz
+1313,FCON1000,fcon1000Milwaukee_sub76378,52.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub76378.nii.gz
+1314,FCON1000,fcon1000Milwaukee_sub77073,52.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub77073.nii.gz
+1315,FCON1000,fcon1000Milwaukee_sub77622,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub77622.nii.gz
+1316,FCON1000,fcon1000Milwaukee_sub78339,,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub78339.nii.gz
+1317,FCON1000,fcon1000Milwaukee_sub78549,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub78549.nii.gz
+1318,FCON1000,fcon1000Milwaukee_sub79672,,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub79672.nii.gz
+1319,FCON1000,fcon1000Milwaukee_sub84314,58.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub84314.nii.gz
+1320,FCON1000,fcon1000Milwaukee_sub86308,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub86308.nii.gz
+1321,FCON1000,fcon1000Milwaukee_sub87784,58.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub87784.nii.gz
+1322,FCON1000,fcon1000Milwaukee_sub87910,47.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub87910.nii.gz
+1323,FCON1000,fcon1000Milwaukee_sub90565,,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub90565.nii.gz
+1324,FCON1000,fcon1000Milwaukee_sub91468,44.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub91468.nii.gz
+1325,FCON1000,fcon1000Milwaukee_sub92023,,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub92023.nii.gz
+1326,FCON1000,fcon1000Milwaukee_sub93170,65.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub93170.nii.gz
+1327,FCON1000,fcon1000Milwaukee_sub98971,63.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub98971.nii.gz
+1328,FCON1000,fcon1000Milwaukee_sub99479,65.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub99479.nii.gz
+1329,FCON1000,fcon1000Milwaukee_sub99558,,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Milwaukee_sub99558.nii.gz
+1330,FCON1000,fcon1000Munchen_sub09035,70.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Munchen_sub09035.nii.gz
+1331,FCON1000,fcon1000Munchen_sub26223,65.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Munchen_sub26223.nii.gz
+1332,FCON1000,fcon1000Munchen_sub26670,64.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Munchen_sub26670.nii.gz
+1333,FCON1000,fcon1000Munchen_sub28697,73.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Munchen_sub28697.nii.gz
+1334,FCON1000,fcon1000Munchen_sub28902,65.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Munchen_sub28902.nii.gz
+1335,FCON1000,fcon1000Munchen_sub31272,74.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Munchen_sub31272.nii.gz
+1336,FCON1000,fcon1000Munchen_sub36052,63.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Munchen_sub36052.nii.gz
+1337,FCON1000,fcon1000Munchen_sub50162,73.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Munchen_sub50162.nii.gz
+1338,FCON1000,fcon1000Munchen_sub64202,67.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Munchen_sub64202.nii.gz
+1339,FCON1000,fcon1000Munchen_sub66933,72.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Munchen_sub66933.nii.gz
+1340,FCON1000,fcon1000Munchen_sub70942,70.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Munchen_sub70942.nii.gz
+1341,FCON1000,fcon1000Munchen_sub72095,71.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Munchen_sub72095.nii.gz
+1342,FCON1000,fcon1000Munchen_sub74607,65.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Munchen_sub74607.nii.gz
+1343,FCON1000,fcon1000Munchen_sub81737,67.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Munchen_sub81737.nii.gz
+1344,FCON1000,fcon1000Munchen_sub96591,73.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Munchen_sub96591.nii.gz
+1345,FCON1000,fcon1000Munchen_sub96752,63.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Munchen_sub96752.nii.gz
+1346,FCON1000,fcon1000NewYork_sub01912,31.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub01912.nii.gz
+1347,FCON1000,fcon1000NewYork_sub02503,9.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub02503.nii.gz
+1348,FCON1000,fcon1000NewYork_sub03951,37.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub03951.nii.gz
+1349,FCON1000,fcon1000NewYork_sub04856,9.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub04856.nii.gz
+1350,FCON1000,fcon1000NewYork_sub05208,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub05208.nii.gz
+1351,FCON1000,fcon1000NewYork_sub07578,20.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub07578.nii.gz
+1352,FCON1000,fcon1000NewYork_sub08595,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub08595.nii.gz
+1353,FCON1000,fcon1000NewYork_sub09539,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub09539.nii.gz
+1354,FCON1000,fcon1000NewYork_sub10011,40.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub10011.nii.gz
+1355,FCON1000,fcon1000NewYork_sub10582,47.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub10582.nii.gz
+1356,FCON1000,fcon1000NewYork_sub12486,40.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub12486.nii.gz
+1357,FCON1000,fcon1000NewYork_sub13384,28.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub13384.nii.gz
+1358,FCON1000,fcon1000NewYork_sub14299,24.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub14299.nii.gz
+1359,FCON1000,fcon1000NewYork_sub14465,47.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub14465.nii.gz
+1360,FCON1000,fcon1000NewYork_sub15213,16.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub15213.nii.gz
+1361,FCON1000,fcon1000NewYork_sub15758,32.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub15758.nii.gz
+1362,FCON1000,fcon1000NewYork_sub16607,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub16607.nii.gz
+1363,FCON1000,fcon1000NewYork_sub17078,34.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub17078.nii.gz
+1364,FCON1000,fcon1000NewYork_sub17109,46.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub17109.nii.gz
+1365,FCON1000,fcon1000NewYork_sub18638,23.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub18638.nii.gz
+1366,FCON1000,fcon1000NewYork_sub19579,46.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub19579.nii.gz
+1367,FCON1000,fcon1000NewYork_sub20676,49.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub20676.nii.gz
+1368,FCON1000,fcon1000NewYork_sub20691,26.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub20691.nii.gz
+1369,FCON1000,fcon1000NewYork_sub20732,11.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub20732.nii.gz
+1370,FCON1000,fcon1000NewYork_sub21212,30.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub21212.nii.gz
+1371,FCON1000,fcon1000NewYork_sub22349,26.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub22349.nii.gz
+1372,FCON1000,fcon1000NewYork_sub22608,23.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub22608.nii.gz
+1373,FCON1000,fcon1000NewYork_sub23844,20.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub23844.nii.gz
+1374,FCON1000,fcon1000NewYork_sub24528,45.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub24528.nii.gz
+1375,FCON1000,fcon1000NewYork_sub26267,15.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub26267.nii.gz
+1376,FCON1000,fcon1000NewYork_sub27123,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub27123.nii.gz
+1377,FCON1000,fcon1000NewYork_sub28795,10.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub28795.nii.gz
+1378,FCON1000,fcon1000NewYork_sub28808,23.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub28808.nii.gz
+1379,FCON1000,fcon1000NewYork_sub29216,29.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub29216.nii.gz
+1380,FCON1000,fcon1000NewYork_sub29353,21.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub29353.nii.gz
+1381,FCON1000,fcon1000NewYork_sub29935,16.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub29935.nii.gz
+1382,FCON1000,fcon1000NewYork_sub30247,24.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub30247.nii.gz
+1383,FCON1000,fcon1000NewYork_sub30623,26.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub30623.nii.gz
+1384,FCON1000,fcon1000NewYork_sub30860,35.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub30860.nii.gz
+1385,FCON1000,fcon1000NewYork_sub31554,24.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub31554.nii.gz
+1386,FCON1000,fcon1000NewYork_sub31671,12.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub31671.nii.gz
+1387,FCON1000,fcon1000NewYork_sub33062,34.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub33062.nii.gz
+1388,FCON1000,fcon1000NewYork_sub33581,49.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub33581.nii.gz
+1389,FCON1000,fcon1000NewYork_sub35262,28.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub35262.nii.gz
+1390,FCON1000,fcon1000NewYork_sub37864,41.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub37864.nii.gz
+1391,FCON1000,fcon1000NewYork_sub38088,16.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub38088.nii.gz
+1392,FCON1000,fcon1000NewYork_sub41546,24.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub41546.nii.gz
+1393,FCON1000,fcon1000NewYork_sub44395,21.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub44395.nii.gz
+1394,FCON1000,fcon1000NewYork_sub44515,25.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub44515.nii.gz
+1395,FCON1000,fcon1000NewYork_sub44979,15.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub44979.nii.gz
+1396,FCON1000,fcon1000NewYork_sub45217,40.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub45217.nii.gz
+1397,FCON1000,fcon1000NewYork_sub46856,8.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub46856.nii.gz
+1398,FCON1000,fcon1000NewYork_sub47087,25.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub47087.nii.gz
+1399,FCON1000,fcon1000NewYork_sub47633,37.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub47633.nii.gz
+1400,FCON1000,fcon1000NewYork_sub48803,37.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub48803.nii.gz
+1401,FCON1000,fcon1000NewYork_sub48830,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub48830.nii.gz
+1402,FCON1000,fcon1000NewYork_sub50559,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub50559.nii.gz
+1403,FCON1000,fcon1000NewYork_sub51309,27.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub51309.nii.gz
+1404,FCON1000,fcon1000NewYork_sub51677,25.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub51677.nii.gz
+1405,FCON1000,fcon1000NewYork_sub53461,27.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub53461.nii.gz
+1406,FCON1000,fcon1000NewYork_sub53710,34.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub53710.nii.gz
+1407,FCON1000,fcon1000NewYork_sub54828,41.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub54828.nii.gz
+1408,FCON1000,fcon1000NewYork_sub56734,50.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub56734.nii.gz
+1409,FCON1000,fcon1000NewYork_sub59796,34.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub59796.nii.gz
+1410,FCON1000,fcon1000NewYork_sub63915,44.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub63915.nii.gz
+1411,FCON1000,fcon1000NewYork_sub69779,40.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub69779.nii.gz
+1412,FCON1000,fcon1000NewYork_sub73035,26.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub73035.nii.gz
+1413,FCON1000,fcon1000NewYork_sub77203,40.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub77203.nii.gz
+1414,FCON1000,fcon1000NewYork_sub77903,24.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub77903.nii.gz
+1415,FCON1000,fcon1000NewYork_sub84371,38.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/NewYork_sub84371.nii.gz
+1416,FCON1000,fcon1000Newark_sub13411,23.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Newark_sub13411.nii.gz
+1417,FCON1000,fcon1000Newark_sub32580,26.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Newark_sub32580.nii.gz
+1418,FCON1000,fcon1000Newark_sub36023,21.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Newark_sub36023.nii.gz
+1419,FCON1000,fcon1000Newark_sub41006,22.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Newark_sub41006.nii.gz
+1420,FCON1000,fcon1000Newark_sub43517,22.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Newark_sub43517.nii.gz
+1421,FCON1000,fcon1000Newark_sub49705,24.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Newark_sub49705.nii.gz
+1422,FCON1000,fcon1000Newark_sub53422,24.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Newark_sub53422.nii.gz
+1423,FCON1000,fcon1000Newark_sub54933,26.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Newark_sub54933.nii.gz
+1424,FCON1000,fcon1000Newark_sub55760,24.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Newark_sub55760.nii.gz
+1425,FCON1000,fcon1000Newark_sub58526,25.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Newark_sub58526.nii.gz
+1426,FCON1000,fcon1000Newark_sub59397,24.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Newark_sub59397.nii.gz
+1427,FCON1000,fcon1000Newark_sub59799,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Newark_sub59799.nii.gz
+1428,FCON1000,fcon1000Newark_sub62985,25.0,dev,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Newark_sub62985.nii.gz
+1429,FCON1000,fcon1000Newark_sub71042,21.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Newark_sub71042.nii.gz
+1430,FCON1000,fcon1000Newark_sub71743,22.0,test,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Newark_sub71743.nii.gz
+1431,FCON1000,fcon1000Newark_sub76647,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Newark_sub76647.nii.gz
+1432,FCON1000,fcon1000Newark_sub78225,22.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Newark_sub78225.nii.gz
+1433,FCON1000,fcon1000Newark_sub86204,23.0,train,/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/Newark_sub86204.nii.gz
+1434,DLBS,DLBS0028374_session_1,55.62,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028374_session_1.nii.gz
+1435,DLBS,DLBS0028545_session_1,72.25,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028545_session_1.nii.gz
+1436,DLBS,DLBS0028344_session_1,69.81,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028344_session_1.nii.gz
+1437,DLBS,DLBS0028392_session_1,54.85,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028392_session_1.nii.gz
+1438,DLBS,DLBS0028479_session_1,40.93,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028479_session_1.nii.gz
+1439,DLBS,DLBS0028495_session_1,40.56,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028495_session_1.nii.gz
+1440,DLBS,DLBS0028492_session_1,52.89,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028492_session_1.nii.gz
+1441,DLBS,DLBS0028468_session_1,44.31,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028468_session_1.nii.gz
+1442,DLBS,DLBS0028592_session_1,40.97,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028592_session_1.nii.gz
+1443,DLBS,DLBS0028419_session_1,78.87,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028419_session_1.nii.gz
+1444,DLBS,DLBS0028550_session_1,22.17,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028550_session_1.nii.gz
+1445,DLBS,DLBS0028632_session_1,59.96,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028632_session_1.nii.gz
+1446,DLBS,DLBS0028352_session_1,29.84,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028352_session_1.nii.gz
+1447,DLBS,DLBS0028366_session_1,23.53,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028366_session_1.nii.gz
+1448,DLBS,DLBS0028429_session_1,42.63,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028429_session_1.nii.gz
+1449,DLBS,DLBS0028456_session_1,53.75,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028456_session_1.nii.gz
+1450,DLBS,DLBS0028593_session_1,62.06,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028593_session_1.nii.gz
+1451,DLBS,DLBS0028549_session_1,21.45,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028549_session_1.nii.gz
+1452,DLBS,DLBS0028617_session_1,50.6,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028617_session_1.nii.gz
+1453,DLBS,DLBS0028340_session_1,67.83,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028340_session_1.nii.gz
+1454,DLBS,DLBS0028349_session_1,64.17,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028349_session_1.nii.gz
+1455,DLBS,DLBS0028505_session_1,20.75,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028505_session_1.nii.gz
+1456,DLBS,DLBS0028463_session_1,68.35,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028463_session_1.nii.gz
+1457,DLBS,DLBS0028603_session_1,35.96,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028603_session_1.nii.gz
+1458,DLBS,DLBS0028441_session_1,45.18,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028441_session_1.nii.gz
+1459,DLBS,DLBS0028428_session_1,45.85,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028428_session_1.nii.gz
+1460,DLBS,DLBS0028523_session_1,49.68,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028523_session_1.nii.gz
+1461,DLBS,DLBS0028357_session_1,56.36,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028357_session_1.nii.gz
+1462,DLBS,DLBS0028574_session_1,50.35,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028574_session_1.nii.gz
+1463,DLBS,DLBS0028499_session_1,51.56,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028499_session_1.nii.gz
+1464,DLBS,DLBS0028569_session_1,67.65,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028569_session_1.nii.gz
+1465,DLBS,DLBS0028385_session_1,61.53,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028385_session_1.nii.gz
+1466,DLBS,DLBS0028381_session_1,24.24,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028381_session_1.nii.gz
+1467,DLBS,DLBS0028631_session_1,29.92,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028631_session_1.nii.gz
+1468,DLBS,DLBS0028500_session_1,32.85,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028500_session_1.nii.gz
+1469,DLBS,DLBS0028326_session_1,66.42,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028326_session_1.nii.gz
+1470,DLBS,DLBS0028628_session_1,51.81,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028628_session_1.nii.gz
+1471,DLBS,DLBS0028461_session_1,53.95,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028461_session_1.nii.gz
+1472,DLBS,DLBS0028560_session_1,57.03,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028560_session_1.nii.gz
+1473,DLBS,DLBS0028532_session_1,34.11,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028532_session_1.nii.gz
+1474,DLBS,DLBS0028520_session_1,65.52,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028520_session_1.nii.gz
+1475,DLBS,DLBS0028528_session_1,53.43,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028528_session_1.nii.gz
+1476,DLBS,DLBS0028438_session_1,85.99,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028438_session_1.nii.gz
+1477,DLBS,DLBS0028424_session_1,77.79,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028424_session_1.nii.gz
+1478,DLBS,DLBS0028600_session_1,70.31,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028600_session_1.nii.gz
+1479,DLBS,DLBS0028636_session_1,65.12,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028636_session_1.nii.gz
+1480,DLBS,DLBS0028368_session_1,79.65,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028368_session_1.nii.gz
+1481,DLBS,DLBS0028388_session_1,82.93,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028388_session_1.nii.gz
+1482,DLBS,DLBS0028333_session_1,76.45,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028333_session_1.nii.gz
+1483,DLBS,DLBS0028413_session_1,81.57,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028413_session_1.nii.gz
+1484,DLBS,DLBS0028615_session_1,80.95,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028615_session_1.nii.gz
+1485,DLBS,DLBS0028457_session_1,75.67,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028457_session_1.nii.gz
+1486,DLBS,DLBS0028399_session_1,70.68,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028399_session_1.nii.gz
+1487,DLBS,DLBS0028339_session_1,65.23,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028339_session_1.nii.gz
+1488,DLBS,DLBS0028639_session_1,74.96,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028639_session_1.nii.gz
+1489,DLBS,DLBS0028474_session_1,83.7,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028474_session_1.nii.gz
+1490,DLBS,DLBS0028480_session_1,40.26,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028480_session_1.nii.gz
+1491,DLBS,DLBS0028616_session_1,61.54,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028616_session_1.nii.gz
+1492,DLBS,DLBS0028488_session_1,41.33,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028488_session_1.nii.gz
+1493,DLBS,DLBS0028537_session_1,62.2,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028537_session_1.nii.gz
+1494,DLBS,DLBS0028358_session_1,66.63,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028358_session_1.nii.gz
+1495,DLBS,DLBS0028378_session_1,69.19,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028378_session_1.nii.gz
+1496,DLBS,DLBS0028557_session_1,75.89,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028557_session_1.nii.gz
+1497,DLBS,DLBS0028396_session_1,67.74,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028396_session_1.nii.gz
+1498,DLBS,DLBS0028490_session_1,63.73,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028490_session_1.nii.gz
+1499,DLBS,DLBS0028590_session_1,70.79,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028590_session_1.nii.gz
+1500,DLBS,DLBS0028587_session_1,72.59,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028587_session_1.nii.gz
+1501,DLBS,DLBS0028377_session_1,64.37,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028377_session_1.nii.gz
+1502,DLBS,DLBS0028518_session_1,71.09,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028518_session_1.nii.gz
+1503,DLBS,DLBS0028407_session_1,86.86,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028407_session_1.nii.gz
+1504,DLBS,DLBS0028612_session_1,67.61,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028612_session_1.nii.gz
+1505,DLBS,DLBS0028542_session_1,69.59,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028542_session_1.nii.gz
+1506,DLBS,DLBS0028562_session_1,24.29,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028562_session_1.nii.gz
+1507,DLBS,DLBS0028509_session_1,51.39,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028509_session_1.nii.gz
+1508,DLBS,DLBS0028365_session_1,22.82,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028365_session_1.nii.gz
+1509,DLBS,DLBS0028558_session_1,24.83,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028558_session_1.nii.gz
+1510,DLBS,DLBS0028599_session_1,62.14,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028599_session_1.nii.gz
+1511,DLBS,DLBS0028578_session_1,87.34,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028578_session_1.nii.gz
+1512,DLBS,DLBS0028336_session_1,80.65,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028336_session_1.nii.gz
+1513,DLBS,DLBS0028434_session_1,70.1,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028434_session_1.nii.gz
+1514,DLBS,DLBS0028403_session_1,68.71,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028403_session_1.nii.gz
+1515,DLBS,DLBS0028471_session_1,67.47,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028471_session_1.nii.gz
+1516,DLBS,DLBS0028477_session_1,60.81,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028477_session_1.nii.gz
+1517,DLBS,DLBS0028580_session_1,60.28,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028580_session_1.nii.gz
+1518,DLBS,DLBS0028448_session_1,69.32,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028448_session_1.nii.gz
+1519,DLBS,DLBS0028605_session_1,63.69,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028605_session_1.nii.gz
+1520,DLBS,DLBS0028421_session_1,67.05,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028421_session_1.nii.gz
+1521,DLBS,DLBS0028362_session_1,60.87,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028362_session_1.nii.gz
+1522,DLBS,DLBS0028354_session_1,75.42,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028354_session_1.nii.gz
+1523,DLBS,DLBS0028602_session_1,29.19,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028602_session_1.nii.gz
+1524,DLBS,DLBS0028360_session_1,54.03,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028360_session_1.nii.gz
+1525,DLBS,DLBS0028527_session_1,23.64,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028527_session_1.nii.gz
+1526,DLBS,DLBS0028596_session_1,36.92,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028596_session_1.nii.gz
+1527,DLBS,DLBS0028554_session_1,48.67,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028554_session_1.nii.gz
+1528,DLBS,DLBS0028515_session_1,50.9,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028515_session_1.nii.gz
+1529,DLBS,DLBS0028625_session_1,60.37,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028625_session_1.nii.gz
+1530,DLBS,DLBS0028586_session_1,79.85,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028586_session_1.nii.gz
+1531,DLBS,DLBS0028572_session_1,75.37,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028572_session_1.nii.gz
+1532,DLBS,DLBS0028335_session_1,71.34,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028335_session_1.nii.gz
+1533,DLBS,DLBS0028577_session_1,65.52,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028577_session_1.nii.gz
+1534,DLBS,DLBS0028401_session_1,54.05,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028401_session_1.nii.gz
+1535,DLBS,DLBS0028408_session_1,20.7,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028408_session_1.nii.gz
+1536,DLBS,DLBS0028412_session_1,20.89,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028412_session_1.nii.gz
+1537,DLBS,DLBS0028394_session_1,20.74,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028394_session_1.nii.gz
+1538,DLBS,DLBS0028425_session_1,23.02,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028425_session_1.nii.gz
+1539,DLBS,DLBS0028375_session_1,59.07,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028375_session_1.nii.gz
+1540,DLBS,DLBS0028439_session_1,25.38,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028439_session_1.nii.gz
+1541,DLBS,DLBS0028476_session_1,65.91,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028476_session_1.nii.gz
+1542,DLBS,DLBS0028519_session_1,64.79,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028519_session_1.nii.gz
+1543,DLBS,DLBS0028444_session_1,81.41,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028444_session_1.nii.gz
+1544,DLBS,DLBS0028555_session_1,64.86,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028555_session_1.nii.gz
+1545,DLBS,DLBS0028379_session_1,73.36,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028379_session_1.nii.gz
+1546,DLBS,DLBS0028498_session_1,67.16,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028498_session_1.nii.gz
+1547,DLBS,DLBS0028414_session_1,68.45,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028414_session_1.nii.gz
+1548,DLBS,DLBS0028369_session_1,62.67,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028369_session_1.nii.gz
+1549,DLBS,DLBS0028576_session_1,32.92,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028576_session_1.nii.gz
+1550,DLBS,DLBS0028470_session_1,43.19,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028470_session_1.nii.gz
+1551,DLBS,DLBS0028380_session_1,83.22,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028380_session_1.nii.gz
+1552,DLBS,DLBS0028343_session_1,61.27,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028343_session_1.nii.gz
+1553,DLBS,DLBS0028565_session_1,80.84,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028565_session_1.nii.gz
+1554,DLBS,DLBS0028494_session_1,74.24,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028494_session_1.nii.gz
+1555,DLBS,DLBS0028409_session_1,70.68,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028409_session_1.nii.gz
+1556,DLBS,DLBS0028611_session_1,46.79,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028611_session_1.nii.gz
+1557,DLBS,DLBS0028510_session_1,51.3,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028510_session_1.nii.gz
+1558,DLBS,DLBS0028570_session_1,63.31,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028570_session_1.nii.gz
+1559,DLBS,DLBS0028584_session_1,51.24,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028584_session_1.nii.gz
+1560,DLBS,DLBS0028566_session_1,57.36,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028566_session_1.nii.gz
+1561,DLBS,DLBS0028521_session_1,39.74,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028521_session_1.nii.gz
+1562,DLBS,DLBS0028450_session_1,44.99,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028450_session_1.nii.gz
+1563,DLBS,DLBS0028497_session_1,49.13,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028497_session_1.nii.gz
+1564,DLBS,DLBS0028398_session_1,83.33,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028398_session_1.nii.gz
+1565,DLBS,DLBS0028331_session_1,77.51,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028331_session_1.nii.gz
+1566,DLBS,DLBS0028400_session_1,49.92,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028400_session_1.nii.gz
+1567,DLBS,DLBS0028589_session_1,59.75,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028589_session_1.nii.gz
+1568,DLBS,DLBS0028447_session_1,51.78,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028447_session_1.nii.gz
+1569,DLBS,DLBS0028638_session_1,70.98,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028638_session_1.nii.gz
+1570,DLBS,DLBS0028460_session_1,37.05,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028460_session_1.nii.gz
+1571,DLBS,DLBS0028432_session_1,53.75,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028432_session_1.nii.gz
+1572,DLBS,DLBS0028561_session_1,49.51,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028561_session_1.nii.gz
+1573,DLBS,DLBS0028613_session_1,74.4,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028613_session_1.nii.gz
+1574,DLBS,DLBS0028389_session_1,33.45,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028389_session_1.nii.gz
+1575,DLBS,DLBS0028420_session_1,81.2,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028420_session_1.nii.gz
+1576,DLBS,DLBS0028543_session_1,81.28,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028543_session_1.nii.gz
+1577,DLBS,DLBS0028351_session_1,83.73,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028351_session_1.nii.gz
+1578,DLBS,DLBS0028345_session_1,59.73,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028345_session_1.nii.gz
+1579,DLBS,DLBS0028376_session_1,70.6,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028376_session_1.nii.gz
+1580,DLBS,DLBS0028541_session_1,45.8,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028541_session_1.nii.gz
+1581,DLBS,DLBS0028622_session_1,24.6,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028622_session_1.nii.gz
+1582,DLBS,DLBS0028530_session_1,49.86,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028530_session_1.nii.gz
+1583,DLBS,DLBS0028548_session_1,40.83,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028548_session_1.nii.gz
+1584,DLBS,DLBS0028332_session_1,71.29,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028332_session_1.nii.gz
+1585,DLBS,DLBS0028427_session_1,25.7,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028427_session_1.nii.gz
+1586,DLBS,DLBS0028350_session_1,49.82,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028350_session_1.nii.gz
+1587,DLBS,DLBS0028404_session_1,38.45,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028404_session_1.nii.gz
+1588,DLBS,DLBS0028508_session_1,28.67,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028508_session_1.nii.gz
+1589,DLBS,DLBS0028387_session_1,54.68,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028387_session_1.nii.gz
+1590,DLBS,DLBS0028591_session_1,86.25,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028591_session_1.nii.gz
+1591,DLBS,DLBS0028346_session_1,57.48,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028346_session_1.nii.gz
+1592,DLBS,DLBS0028512_session_1,51.64,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028512_session_1.nii.gz
+1593,DLBS,DLBS0028633_session_1,46.44,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028633_session_1.nii.gz
+1594,DLBS,DLBS0028582_session_1,48.45,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028582_session_1.nii.gz
+1595,DLBS,DLBS0028598_session_1,84.06,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028598_session_1.nii.gz
+1596,DLBS,DLBS0028571_session_1,20.57,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028571_session_1.nii.gz
+1597,DLBS,DLBS0028337_session_1,53.34,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028337_session_1.nii.gz
+1598,DLBS,DLBS0028431_session_1,47.63,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028431_session_1.nii.gz
+1599,DLBS,DLBS0028442_session_1,29.32,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028442_session_1.nii.gz
+1600,DLBS,DLBS0028373_session_1,48.4,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028373_session_1.nii.gz
+1601,DLBS,DLBS0028493_session_1,53.36,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028493_session_1.nii.gz
+1602,DLBS,DLBS0028475_session_1,54.05,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028475_session_1.nii.gz
+1603,DLBS,DLBS0028635_session_1,23.5,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028635_session_1.nii.gz
+1604,DLBS,DLBS0028363_session_1,26.02,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028363_session_1.nii.gz
+1605,DLBS,DLBS0028559_session_1,25.01,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028559_session_1.nii.gz
+1606,DLBS,DLBS0028534_session_1,54.3,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028534_session_1.nii.gz
+1607,DLBS,DLBS0028383_session_1,72.96,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028383_session_1.nii.gz
+1608,DLBS,DLBS0028496_session_1,22.44,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028496_session_1.nii.gz
+1609,DLBS,DLBS0028418_session_1,70.49,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028418_session_1.nii.gz
+1610,DLBS,DLBS0028579_session_1,21.77,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028579_session_1.nii.gz
+1611,DLBS,DLBS0028367_session_1,29.58,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028367_session_1.nii.gz
+1612,DLBS,DLBS0028624_session_1,81.96,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028624_session_1.nii.gz
+1613,DLBS,DLBS0028533_session_1,46.77,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028533_session_1.nii.gz
+1614,DLBS,DLBS0028539_session_1,24.22,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028539_session_1.nii.gz
+1615,DLBS,DLBS0028575_session_1,81.02,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028575_session_1.nii.gz
+1616,DLBS,DLBS0028629_session_1,81.12,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028629_session_1.nii.gz
+1617,DLBS,DLBS0028489_session_1,70.28,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028489_session_1.nii.gz
+1618,DLBS,DLBS0028640_session_1,33.41,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028640_session_1.nii.gz
+1619,DLBS,DLBS0028620_session_1,70.66,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028620_session_1.nii.gz
+1620,DLBS,DLBS0028459_session_1,66.13,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028459_session_1.nii.gz
+1621,DLBS,DLBS0028484_session_1,70.08,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028484_session_1.nii.gz
+1622,DLBS,DLBS0028348_session_1,23.15,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028348_session_1.nii.gz
+1623,DLBS,DLBS0028517_session_1,23.14,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028517_session_1.nii.gz
+1624,DLBS,DLBS0028604_session_1,22.6,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028604_session_1.nii.gz
+1625,DLBS,DLBS0028607_session_1,66.74,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028607_session_1.nii.gz
+1626,DLBS,DLBS0028327_session_1,73.29,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028327_session_1.nii.gz
+1627,DLBS,DLBS0028601_session_1,71.54,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028601_session_1.nii.gz
+1628,DLBS,DLBS0028370_session_1,22.88,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028370_session_1.nii.gz
+1629,DLBS,DLBS0028507_session_1,36.91,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028507_session_1.nii.gz
+1630,DLBS,DLBS0028361_session_1,55.56,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028361_session_1.nii.gz
+1631,DLBS,DLBS0028618_session_1,26.36,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028618_session_1.nii.gz
+1632,DLBS,DLBS0028449_session_1,28.96,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028449_session_1.nii.gz
+1633,DLBS,DLBS0028522_session_1,34.99,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028522_session_1.nii.gz
+1634,DLBS,DLBS0028514_session_1,34.44,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028514_session_1.nii.gz
+1635,DLBS,DLBS0028531_session_1,21.27,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028531_session_1.nii.gz
+1636,DLBS,DLBS0028551_session_1,24.66,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028551_session_1.nii.gz
+1637,DLBS,DLBS0028621_session_1,30.6,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028621_session_1.nii.gz
+1638,DLBS,DLBS0028513_session_1,26.68,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028513_session_1.nii.gz
+1639,DLBS,DLBS0028454_session_1,42.92,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028454_session_1.nii.gz
+1640,DLBS,DLBS0028411_session_1,83.09,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028411_session_1.nii.gz
+1641,DLBS,DLBS0028546_session_1,21.71,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028546_session_1.nii.gz
+1642,DLBS,DLBS0028538_session_1,76.39,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028538_session_1.nii.gz
+1643,DLBS,DLBS0028364_session_1,24.49,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028364_session_1.nii.gz
+1644,DLBS,DLBS0028390_session_1,22.5,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028390_session_1.nii.gz
+1645,DLBS,DLBS0028359_session_1,71.2,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028359_session_1.nii.gz
+1646,DLBS,DLBS0028334_session_1,55.26,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028334_session_1.nii.gz
+1647,DLBS,DLBS0028430_session_1,89.11,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028430_session_1.nii.gz
+1648,DLBS,DLBS0028487_session_1,73.56,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028487_session_1.nii.gz
+1649,DLBS,DLBS0028563_session_1,44.97,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028563_session_1.nii.gz
+1650,DLBS,DLBS0028445_session_1,55.37,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028445_session_1.nii.gz
+1651,DLBS,DLBS0028453_session_1,73.26,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028453_session_1.nii.gz
+1652,DLBS,DLBS0028481_session_1,41.87,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028481_session_1.nii.gz
+1653,DLBS,DLBS0028465_session_1,21.67,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028465_session_1.nii.gz
+1654,DLBS,DLBS0028511_session_1,23.55,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028511_session_1.nii.gz
+1655,DLBS,DLBS0028443_session_1,36.18,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028443_session_1.nii.gz
+1656,DLBS,DLBS0028568_session_1,41.17,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028568_session_1.nii.gz
+1657,DLBS,DLBS0028552_session_1,48.86,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028552_session_1.nii.gz
+1658,DLBS,DLBS0028614_session_1,53.77,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028614_session_1.nii.gz
+1659,DLBS,DLBS0028417_session_1,31.77,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028417_session_1.nii.gz
+1660,DLBS,DLBS0028626_session_1,48.88,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028626_session_1.nii.gz
+1661,DLBS,DLBS0028393_session_1,61.69,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028393_session_1.nii.gz
+1662,DLBS,DLBS0028573_session_1,44.29,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028573_session_1.nii.gz
+1663,DLBS,DLBS0028564_session_1,40.21,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028564_session_1.nii.gz
+1664,DLBS,DLBS0028451_session_1,43.61,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028451_session_1.nii.gz
+1665,DLBS,DLBS0028415_session_1,32.05,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028415_session_1.nii.gz
+1666,DLBS,DLBS0028585_session_1,31.4,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028585_session_1.nii.gz
+1667,DLBS,DLBS0028386_session_1,33.13,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028386_session_1.nii.gz
+1668,DLBS,DLBS0028391_session_1,36.03,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028391_session_1.nii.gz
+1669,DLBS,DLBS0028467_session_1,34.4,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028467_session_1.nii.gz
+1670,DLBS,DLBS0028384_session_1,30.5,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028384_session_1.nii.gz
+1671,DLBS,DLBS0028328_session_1,30.47,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028328_session_1.nii.gz
+1672,DLBS,DLBS0028482_session_1,63.47,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028482_session_1.nii.gz
+1673,DLBS,DLBS0028353_session_1,60.01,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028353_session_1.nii.gz
+1674,DLBS,DLBS0028544_session_1,81.75,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028544_session_1.nii.gz
+1675,DLBS,DLBS0028469_session_1,50.46,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028469_session_1.nii.gz
+1676,DLBS,DLBS0028491_session_1,24.2,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028491_session_1.nii.gz
+1677,DLBS,DLBS0028504_session_1,73.07,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028504_session_1.nii.gz
+1678,DLBS,DLBS0028506_session_1,86.38,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028506_session_1.nii.gz
+1679,DLBS,DLBS0028355_session_1,70.7,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028355_session_1.nii.gz
+1680,DLBS,DLBS0028356_session_1,43.21,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028356_session_1.nii.gz
+1681,DLBS,DLBS0028455_session_1,72.56,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028455_session_1.nii.gz
+1682,DLBS,DLBS0028436_session_1,75.99,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028436_session_1.nii.gz
+1683,DLBS,DLBS0028478_session_1,70.35,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028478_session_1.nii.gz
+1684,DLBS,DLBS0028529_session_1,86.77,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028529_session_1.nii.gz
+1685,DLBS,DLBS0028466_session_1,74.5,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028466_session_1.nii.gz
+1686,DLBS,DLBS0028462_session_1,31.03,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028462_session_1.nii.gz
+1687,DLBS,DLBS0028452_session_1,59.71,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028452_session_1.nii.gz
+1688,DLBS,DLBS0028588_session_1,48.53,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028588_session_1.nii.gz
+1689,DLBS,DLBS0028525_session_1,60.74,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028525_session_1.nii.gz
+1690,DLBS,DLBS0028486_session_1,37.81,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028486_session_1.nii.gz
+1691,DLBS,DLBS0028371_session_1,31.11,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028371_session_1.nii.gz
+1692,DLBS,DLBS0028483_session_1,47.0,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028483_session_1.nii.gz
+1693,DLBS,DLBS0028634_session_1,38.33,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028634_session_1.nii.gz
+1694,DLBS,DLBS0028608_session_1,38.13,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028608_session_1.nii.gz
+1695,DLBS,DLBS0028458_session_1,35.43,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028458_session_1.nii.gz
+1696,DLBS,DLBS0028416_session_1,31.58,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028416_session_1.nii.gz
+1697,DLBS,DLBS0028329_session_1,30.74,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028329_session_1.nii.gz
+1698,DLBS,DLBS0028338_session_1,30.26,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028338_session_1.nii.gz
+1699,DLBS,DLBS0028341_session_1,34.81,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028341_session_1.nii.gz
+1700,DLBS,DLBS0028535_session_1,80.11,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028535_session_1.nii.gz
+1701,DLBS,DLBS0028553_session_1,30.76,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028553_session_1.nii.gz
+1702,DLBS,DLBS0028630_session_1,38.37,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028630_session_1.nii.gz
+1703,DLBS,DLBS0028426_session_1,82.07,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028426_session_1.nii.gz
+1704,DLBS,DLBS0028623_session_1,30.31,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028623_session_1.nii.gz
+1705,DLBS,DLBS0028609_session_1,85.77,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028609_session_1.nii.gz
+1706,DLBS,DLBS0028502_session_1,82.54,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028502_session_1.nii.gz
+1707,DLBS,DLBS0028423_session_1,30.69,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028423_session_1.nii.gz
+1708,DLBS,DLBS0028627_session_1,47.3,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028627_session_1.nii.gz
+1709,DLBS,DLBS0028524_session_1,56.55,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028524_session_1.nii.gz
+1710,DLBS,DLBS0028402_session_1,86.41,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028402_session_1.nii.gz
+1711,DLBS,DLBS0028464_session_1,82.43,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028464_session_1.nii.gz
+1712,DLBS,DLBS0028595_session_1,87.68,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028595_session_1.nii.gz
+1713,DLBS,DLBS0028372_session_1,74.83,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028372_session_1.nii.gz
+1714,DLBS,DLBS0028330_session_1,35.36,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028330_session_1.nii.gz
+1715,DLBS,DLBS0028619_session_1,36.41,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028619_session_1.nii.gz
+1716,DLBS,DLBS0028485_session_1,57.23,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028485_session_1.nii.gz
+1717,DLBS,DLBS0028440_session_1,34.25,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028440_session_1.nii.gz
+1718,DLBS,DLBS0028536_session_1,82.57,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028536_session_1.nii.gz
+1719,DLBS,DLBS0028410_session_1,74.11,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028410_session_1.nii.gz
+1720,DLBS,DLBS0028437_session_1,64.01,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028437_session_1.nii.gz
+1721,DLBS,DLBS0028501_session_1,88.36,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028501_session_1.nii.gz
+1722,DLBS,DLBS0028473_session_1,56.11,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028473_session_1.nii.gz
+1723,DLBS,DLBS0028347_session_1,85.66,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028347_session_1.nii.gz
+1724,DLBS,DLBS0028594_session_1,80.29,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028594_session_1.nii.gz
+1725,DLBS,DLBS0028606_session_1,81.91,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028606_session_1.nii.gz
+1726,DLBS,DLBS0028382_session_1,88.07,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028382_session_1.nii.gz
+1727,DLBS,DLBS0028583_session_1,80.54,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028583_session_1.nii.gz
+1728,DLBS,DLBS0028516_session_1,88.14,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028516_session_1.nii.gz
+1729,DLBS,DLBS0028405_session_1,86.45,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028405_session_1.nii.gz
+1730,DLBS,DLBS0028395_session_1,83.87,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028395_session_1.nii.gz
+1731,DLBS,DLBS0028503_session_1,88.32,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028503_session_1.nii.gz
+1732,DLBS,DLBS0028397_session_1,35.33,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028397_session_1.nii.gz
+1733,DLBS,DLBS0028610_session_1,80.82,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028610_session_1.nii.gz
+1734,DLBS,DLBS0028446_session_1,50.89,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028446_session_1.nii.gz
+1735,DLBS,DLBS0028433_session_1,80.5,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028433_session_1.nii.gz
+1736,DLBS,DLBS0028435_session_1,33.01,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028435_session_1.nii.gz
+1737,DLBS,DLBS0028597_session_1,52.52,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028597_session_1.nii.gz
+1738,DLBS,DLBS0028581_session_1,74.58,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028581_session_1.nii.gz
+1739,DLBS,DLBS0028567_session_1,85.6,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028567_session_1.nii.gz
+1740,DLBS,DLBS0028540_session_1,71.04,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028540_session_1.nii.gz
+1741,DLBS,DLBS0028342_session_1,31.98,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028342_session_1.nii.gz
+1742,DLBS,DLBS0028637_session_1,44.54,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028637_session_1.nii.gz
+1743,DLBS,DLBS0028526_session_1,42.6,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028526_session_1.nii.gz
+1744,DLBS,DLBS0028472_session_1,42.45,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028472_session_1.nii.gz
+1745,DLBS,DLBS0028547_session_1,32.13,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028547_session_1.nii.gz
+1746,DLBS,DLBS0028406_session_1,47.65,train,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028406_session_1.nii.gz
+1747,DLBS,DLBS0028556_session_1,36.97,test,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028556_session_1.nii.gz
+1748,DLBS,DLBS0028422_session_1,32.91,dev,/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/0028422_session_1.nii.gz
diff --git a/create_ds.py b/create_ds.py
new file mode 100644
index 0000000..fa9b452
--- /dev/null
+++ b/create_ds.py
@@ -0,0 +1,123 @@
+import pandas as pd
+import numpy as np
+import os
+
+
+# Load the metadata CSV
+df_ixi = pd.read_csv("ixi_storage/ixi_brainrotnet_metadata.csv")
+
+
+df_ixi = pd.read_csv("ixi_storage/ixi_brainrotnet_metadata.csv")
+df_ixi['uid'] = df_ixi['ImageID']
+
+# Directory containing the files
+ixi_dir = "ixi_storage/IXI-T1"
+
+# List all files in the directory and get their absolute paths
+file_list = [os.path.abspath(os.path.join(ixi_dir, f)) for f in os.listdir(ixi_dir) if os.path.isfile(os.path.join(ixi_dir, f))]
+
+# Function to find the matching file path based on the first 6 characters
+def find_matching_path(image_id, files):
+    for file_path in files:
+        file_name = os.path.basename(file_path)
+        if file_name[:6] == image_id:
+            return file_path
+    return None
+
+
+
+# Add the 'path' column by matching ImageID with file names
+df_ixi['path'] = df_ixi['ImageID'].apply(lambda x: find_matching_path(x, file_list))
+
+# Display a sample of the updated DataFrame
+# Drop columns 'Sex' and 'ImageID'
+df_ixi = df_ixi.drop(columns=['Sex', 'ImageID'])
+
+# Rename the column 'Age' to 'age_at_scan'
+df_ixi = df_ixi.rename(columns={'Age': 'age_at_scan'})
+
+# print head of the updated DataFrame
+# print(df_ixi.head())# Load the metadata CSV
+df_ixi['Project'] = 'IXI'
+
+
+df_fcon = pd.read_csv("fcon_storage/fcon1000_brainrotnet_metadata.csv")
+df_fcon['uid'] = df_fcon['ImageID']
+#drop first 8 characters from the 'ImageID' column, ad .nii.gz to the end and add     /home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/ to the beginning
+df_fcon['ImageID'] = '/home/nafiu/BasicData/cse472_DL_project/fcon_storage/fcon1000/' + df_fcon['ImageID'].str[8:] + '.nii.gz'
+df_fcon = df_fcon.rename(columns={'ImageID': 'path'})
+df_fcon = df_fcon.rename(columns={'Age': 'age_at_scan'})
+df_fcon = df_fcon.drop(columns=['Sex'])
+df_fcon['Project'] = 'FCON1000'
+
+# print(df_fcon.head())# Load the metadata CSV
+
+# new csv save
+df_ixi.to_csv('ixi_storage/ixi_metadata.csv', index=False)
+
+
+df_dlbs = pd.read_csv("dlbs_storage/dlbs_brainrotnet_metadata.csv")
+df_dlbs['uid'] = df_dlbs['ImageID']
+df_dlbs['ImageID'] = '/home/nafiu/BasicData/cse472_DL_project/dlbs_storage/DLBS/' + df_dlbs['ImageID'].str[4:] + '.nii.gz'
+
+df_dlbs = df_dlbs.rename(columns={'ImageID': 'path'})
+df_dlbs = df_dlbs.rename(columns={'Age': 'age_at_scan'})
+df_dlbs = df_dlbs.drop(columns=['Sex'])
+df_dlbs['Project'] = 'DLBS'
+
+
+df_dlbs.to_csv('dlbs_storage/dlbs_metadata.csv', index=False)
+
+# combine all the dataframes
+df_combined = pd.concat([df_ixi, df_fcon, df_dlbs], ignore_index=True)
+# save the combined dataframe to a csv file
+df_combined.to_csv('combined_metadata.csv', index=False)
+df_combined = pd.read_csv("combined_metadata.csv")  # Optionally use 'index_col=0' if the first column is the index
+print(df_combined.head())
+
+import uuid
+
+# Assuming df_combined is already loaded
+# Example: df_combined = pd.read_csv("combined_metadata.csv")
+
+# Step 1: Add a unique 'indx' column (starting from 0)
+df_combined['indx'] = range(len(df_combined))
+
+# Step 2: Add a unique 'uid' column using UUIDs
+
+# Step 3: Randomly assign 'train', 'dev', or 'test' based on the desired distribution
+# Create a list of partitions: 60% train, 20% dev, 20% test
+np.random.seed(42)  # For reproducibility
+partition_labels = ['train'] * int(0.6 * len(df_combined)) + \
+                   ['dev']   * int(0.2 * len(df_combined)) + \
+                   ['test']  * int(0.2 * len(df_combined))
+
+# If there are any remaining rows due to rounding errors, add them as 'train'
+remaining = len(df_combined) - len(partition_labels)
+partition_labels += ['train'] * remaining
+
+# Shuffle the partition labels randomly
+np.random.shuffle(partition_labels)
+
+# Assign the shuffled partition labels to the DataFrame
+df_combined['partition'] = partition_labels
+
+# Save the updated DataFrame (optional)
+df_combined.to_csv("combined_metadata.csv", index=False)
+
+# Display the first few rows
+print(df_combined.head())
+print(df_combined.columns)
+
+# Define the desired column order
+desired_order = ['indx', 'Project', 'uid', 'age_at_scan', 'partition', 'path']
+
+# Reorder the columns
+df_combined = df_combined[desired_order]
+
+# Save the updated DataFrame to a CSV (optional)
+df_combined.to_csv("combined_metadata.csv", index=False)
+
+# Verify the new column order
+print(df_combined.head())
+print(df_combined.columns)
\ No newline at end of file
diff --git a/cringe.py b/cringe.py
index e60d0c8..8ff8dd7 100644
--- a/cringe.py
+++ b/cringe.py
@@ -1,68 +1,443 @@
+import os
+import pandas as pd
+import nibabel as nib
+import numpy as np
+import torch
+import torch.nn as nn
+import torch.nn.init as init
+import torch.optim as optim
+from tqdm import tqdm
+from nibabel.orientations import io_orientation, axcodes2ornt, ornt_transform, apply_orientation
+from torchvision import transforms
+from transformers import ViTFeatureExtractor, ViTModel
+from transformers import ViTForImageClassification
+from transformers import DeiTForImageClassification
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler
+import pickle
+import matplotlib.pyplot as plt
+import sys
+import SimpleITK as sitk
+from scipy.ndimage import zoom
+from dataset_cls import ADNIDataset, ADNIDatasetViT
+from torch.utils.data import DataLoader, Dataset
+import gc
+
+def set_random_seed(seed=69420):
+    np.random.seed(seed)
+    torch.manual_seed(seed)
+    torch.cuda.manual_seed(seed)
+    torch.cuda.manual_seed_all(seed)  # For multi-GPU setups
+    torch.backends.cudnn.deterministic = True
+    torch.backends.cudnn.benchmark = False
+
+universal_seed = 69420
+
+set_random_seed(universal_seed)
+
+def resample_nifti(img_data, target_slices = 160):
+    # Determine the current number of slices along the z-axis (3rd dimension)
+    current_slices = img_data.shape[0]
+    # Calculate the zoom factor for resampling (only along the z-axis)
+    zoom_factor = target_slices / current_slices
+    # Resample the image data along the z-axis
+    resampled_data = zoom(img_data, (zoom_factor, 1, 1), order=3)  # order=3 for cubic interpolation
+    return resampled_data
+
+
+# Check if GPU is available
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 
 # Load the CSV file into a pandas DataFrame
 csv_path = "adni_storage/adni_brainrotnet_metadata.csv"
-df = pd.read_csv(csv_path)
+df_adni = pd.read_csv(csv_path)
 # df = df.sample(n=1000, random_state=69420)
-print (df)
 # Add a new column 'filepath' with the constructed file paths
-df['filepath'] = df.apply(
-    lambda row: f"adni_storage/ADNI_nii_gz_bias_corrected/I{row['ImageID']}_{row['SubjectID']}.stripped.N4.nii.gz",
+df_adni['filepath'] = df_adni.apply(
+    lambda row: f"adni_storage/ADNI_nii_gz_bias_corrected/I{row['ImageID'][4:]}_{row['SubjectID']}.stripped.N4.nii.gz",
     axis=1
 )
+df_adni = df_adni.sort_values(by='Age', ascending=True).reset_index(drop=True).head(500)
+# df_adni=df_adni.sample(n=400)
 
+df = pd.concat ([
+                 df_adni[['ImageID', 'Sex', 'Age', 'filepath']], 
+                
+                 ], ignore_index=True)
 
-# Prepare dataset and dataloaders
+df['Age_Group'] = df['Age'].astype(int).apply(lambda x: f"{x:03d}"[:-1] + "0")
+df['Age_Group'] = df['Age_Group'] + df['Sex']
+print (df['Age_Group'].unique())
 sex_encoded = df['Sex'].apply(lambda x: 0 if x == 'M' else 1).tolist()
 age_list = df['Age'].tolist()
+filepath_list = df['filepath'].tolist()
+label_list = df['Age_Group'].tolist()
+
+unique_labels = sorted(set(label_list))  # Ensure consistent ordering
+label_to_idx = {label: idx for idx, label in enumerate(unique_labels)}
+idx_to_label = {idx: label for label, idx in label_to_idx.items()}  # Reverse mapping for decoding
+
+# Convert labels to integers
+numeric_labels = [label_to_idx[label] for label in label_list]
+label_list = numeric_labels
+
+roi = 160
+
+# Transformation pipeline for ViT
+transform = transforms.Compose([
+    transforms.ToPILImage(),
+    transforms.Resize((224, 224)),
+    transforms.Lambda(lambda img: img.convert("RGB")),  # Convert to RGB
+    transforms.ToTensor(),
+    transforms.Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5]),  # Normalize for ViT
+])
+
+
+# Function to extract 16 evenly spaced slices
+def extract_slices(volume, num_slices=16):
+    total_slices = volume.shape[0]
+    indices = np.linspace(0, total_slices - 1, num_slices, dtype=int)
+    return volume[indices, :, :]  # Select slices
+
+def calculate_bounding_box_from_volume(volume):
+    # Find indices of non-zero values
+    non_zero_indices = np.argwhere(volume > 0)
+
+    # Calculate min and max indices along each dimension
+    min_indices = np.min(non_zero_indices, axis=0)
+    max_indices = np.max(non_zero_indices, axis=0)
+
+    # Convert indices to integers
+    min_indices = min_indices.astype(int)
+    max_indices = max_indices.astype(int)
+
+    return min_indices, max_indices
+
+
+def crop_brain_volumes(brain_data):
+    
+
+        # Calculate bounding box from the brain volume
+    min_indices, max_indices = calculate_bounding_box_from_volume(brain_data)
+
+        # Crop the volume
+    cropped_brain = brain_data[min_indices[0]:max_indices[0] + 1,
+                                   min_indices[1]:max_indices[1] + 1,
+                                   min_indices[2]:max_indices[2] + 1]
+    return cropped_brain
+
+# Function to preprocess data and dynamically expand slices while saving to disk
+def preprocess_and_expand(dataset, transform, output_dir, num_slices=16):
+    os.makedirs(output_dir, exist_ok=True)  # Ensure output directory exists
+    expanded_images, expanded_labels = [], []
+
+    for filepath, label in tqdm(dataset, desc="Processing Slices"):
+        # Check if all slice files already exist
+        all_slices_exist = True
+        slice_filenames = [
+            os.path.join(output_dir, f"{os.path.basename(filepath)}_slice_{i}.pt")
+            for i in range(num_slices)
+        ]
+        if not all(os.path.exists(slice_file) for slice_file in slice_filenames):
+            all_slices_exist = False
+
+        # Skip processing if all slices exist
+        if all_slices_exist:
+            expanded_images.extend(slice_filenames)  # Add existing file paths
+            expanded_labels.extend([label] * num_slices)
+            continue
+
+        # Load NIfTI image only if slices are missing
+        nii_img = nib.load(filepath)
+        orig_ornt = io_orientation(nii_img.affine)
+        ras_ornt = axcodes2ornt(("R", "A", "S"))
+        ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+        data = nii_img.get_fdata()  # Load image data
+        data = apply_orientation(data, ornt_trans)
+
+        data = crop_brain_volumes(data)
+
+        # Normalize and extract slices
+        data = (data - data.min()) / (data.max() - data.min())
+        slices = extract_slices(data, num_slices)
+
+        # Transform each slice, save to file, and add to dataset
+        for i, slice_data in enumerate(slices):
+            slice_filename = slice_filenames[i]
+            if not os.path.exists(slice_filename):
+                transformed_slice = transform(slice_data)  # Transform slice
+                torch.save(transformed_slice, slice_filename)  # Save to file
+            expanded_images.append(slice_filename)  # Store file path
+            expanded_labels.append(label)
+
+    return expanded_images, expanded_labels
+# Instantiate Dataset
+vit_dataset = ADNIDatasetViT(filepath_list, label_list)
+
+# Split Dataset
+train_size = int(0.8 * len(vit_dataset))
+val_size = len(vit_dataset) - train_size
+generator = torch.Generator().manual_seed(universal_seed)
+vit_train_dataset, vit_val_dataset = torch.utils.data.random_split(vit_dataset, [train_size, val_size], generator=generator)
+
+# Create New Dataset with Filepaths
+class ExpandedDataset(Dataset):
+    def __init__(self, image_paths, labels):
+        self.image_paths = image_paths
+        self.labels = labels
+
+    def __len__(self):
+        return len(self.labels)
+
+    def __getitem__(self, idx):
+        # Load the image from file
+        image = torch.load(self.image_paths[idx])
+        label = torch.tensor(self.labels[idx], dtype=torch.long)
+        return image, label
+
+slice_count = 32
+
+# Define output directory for slices
+output_dir = f"processed_slices/{slice_count}"
+
+# Preprocess and expand the training data
+expanded_image_paths, expanded_labels = preprocess_and_expand(vit_train_dataset, transform, output_dir, num_slices=slice_count)
+
+# Create Expanded Dataset and DataLoader
+expanded_train_dataset = ExpandedDataset(expanded_image_paths, expanded_labels)
+expanded_train_loader = DataLoader(expanded_train_dataset, batch_size=8, shuffle=True)
+
+
+# Load ViT model
+num_classes = df['Age_Group'].nunique()  # Number of unique Age_Groups
+model = ViTForImageClassification.from_pretrained(
+    "google/vit-base-patch16-224",
+    num_labels=num_classes,
+    ignore_mismatched_sizes=True, 
+)
+
+model.to(device)
+
+# Loss function and optimizer
+criterion = nn.CrossEntropyLoss()
+optimizer = optim.AdamW(model.parameters(), lr=1e-4)
+
+# Function to save checkpoint
+def save_checkpoint(epoch, model, optimizer, path=f"model_dumps/vit_train_checkpoint_{slice_count}.pth"):
+    torch.save({
+        'epoch': epoch,
+        'model_state_dict': model.state_dict(),
+        'optimizer_state_dict': optimizer.state_dict(),
+    }, path)
+    print(f"Checkpoint saved at epoch {epoch+1}")
+
+vit_train_epochs = 6
+model.train()
+start_epoch = 0
+
+for epoch in range(start_epoch, vit_train_epochs):
+    running_loss = 0.0
+    correct = 0
+    total = 0
+
+    for inputs, labels in tqdm(expanded_train_loader, desc=f"Epoch {epoch+1}/{vit_train_epochs}"):
+        # Move data to device
+        inputs, labels = inputs.to(device), labels.to(device)
+
+        outputs = model(pixel_values=inputs)  # ViT expects `pixel_values`
+        loss = criterion(outputs.logits, labels)
+
+        optimizer.zero_grad()
+        loss.backward()
+        optimizer.step()
+
+        running_loss += loss.item()
+
+        _, predicted = torch.max(outputs.logits, dim=1)
+        total += labels.size(0)
+        correct += (predicted == labels).sum().item()
+
+    epoch_loss = running_loss / len(expanded_train_loader)
+    epoch_accuracy = 100 * correct / total
+    print(f"Epoch {epoch+1}/{vit_train_epochs}, Loss: {epoch_loss:.4f}, Accuracy: {epoch_accuracy:.2f}%")
+    # sav the accuracy and loss for each epoch in a csv file
+    with open(f'vit_train_metrics_{slice_count}.csv', 'a') as f:
+        f.write(f"{epoch+1},{epoch_loss},{epoch_accuracy}\n")
+    
+    save_checkpoint(epoch, model, optimizer, path=f"model_dumps/vit_train_checkpoint_{slice_count}.pth")
+    # model.to(device)  # Move back to GPU
+    gc.collect()
+
+feature_extractor = ViTFeatureExtractor.from_pretrained("google/vit-base-patch16-224")
+model.eval()
+
+transform = transforms.Compose([
+    transforms.ToPILImage(),
+    transforms.Resize((224, 224)),
+    transforms.Lambda(lambda img: img.convert("RGB")),  # Convert to RGB directly
+    transforms.ToTensor(),
+    transforms.Normalize(mean=feature_extractor.image_mean, std=feature_extractor.image_std),
+])
+
+torch.cuda.empty_cache()  # Free GPU memory
+
+# To store features and labels
+features_list = []
+labels_list = []
+
+os.makedirs(f"adni_storage/ADNI_features/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+# Process each row in the DataFrame
+for _, row in tqdm(df_adni.iterrows(), total=len(df_adni), desc="Processing images"):
+    filepath = row['filepath']
+    image_title = f"{row['ImageID'][4:]}_{row['SubjectID']}"
+
+    # Check if the feature file already exists
+    feature_file_path = f"adni_storage/ADNI_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        
+        from PIL import Image
+        # Normalize the array to 0-255 for grayscale image
+        data_normalized = ((features - np.min(features)) / (np.max(features) - np.min(features)) * 255).astype(np.uint8)
+        data_normalized = np.repeat(data_normalized, 4, axis=0)
+        # Create an image from the array
+        img = Image.fromarray(np.transpose(data_normalized), mode='L')  # 'L' mode for grayscale
+        # Save the image
+        # img.save(f"adni_storage/ADNI_features/train_e{vit_train_epochs}/{slice_count}/featuremaps/{image_title}_fm.png")
+
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        # print ("hiii")
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                data = crop_brain_volumes(data)
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+
+                # Extract features for all sagittal slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+
+                    # Transform slice for ViT input
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)  # Add batch dimension and move to GPU
+
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        # #outputs = model(slice_tensor)
+                        # slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()  # Move output back to CPU
+                        slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Target is 'Age'
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+
+batch_size = 1
 
-# print (features_list)
 print (features_list[0].shape)
 
-# Create Dataset and DataLoader
 dataset = ADNIDataset(features_list, sex_encoded, age_list)
 train_size = int(0.8 * len(dataset))
 val_size = len(dataset) - train_size
-train_dataset, val_dataset = torch.utils.data.random_split(dataset, [train_size, val_size])
-
-# Store the indices of the validation dataset
+generator.manual_seed(universal_seed)
+train_dataset, val_dataset = torch.utils.data.random_split(dataset, [train_size, val_size], generator=generator)
 val_indices = val_dataset.indices
 train_indices = train_dataset.indices
 
-train_loader = DataLoader(train_dataset, batch_size=1, shuffle=True)
-val_loader = DataLoader(val_dataset, batch_size=1, shuffle=False)
-
-
-# Tracking outputs for validation samples
+train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=False)
+val_loader = DataLoader(val_dataset, batch_size=batch_size, shuffle=False)
 val_outputs = {}
 train_outputs = {}
 
-# Initialize model, loss, and optimizer
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 
-model = AgePredictionCNN(features_list[0].shape).to(device)
-criterion = nn.L1Loss()  # MAE Loss
-optimizer = optim.Adam(model.parameters(), lr=0.001)
+
+import importlib
+
+# Assuming sys.argv[1] is the module name
+module_name = sys.argv[1]  # Example: "my_model"
+class_name = "AgePredictionCNN"  # The class you want to import
+
+try:
+    # Dynamically import the module
+    module = importlib.import_module(module_name)
+    
+    # Dynamically get the class
+    AgePredictionCNN = getattr(module, class_name)
+    
+    print(f"Successfully imported {class_name} from {module_name}.")
+
+except ImportError:
+    print(f"Module {module_name} could not be imported.")
+except AttributeError:
+    print(f"{class_name} does not exist in {module_name}.")
+
+##############################
+# MODEL IMPORTED DYNAMICALLY
+##############################
+
+print (features_list[0].shape)
+model = AgePredictionCNN((1, features_list[0].shape[0], features_list[0].shape[1])).to(device)
+criterion = nn.L1Loss()
+eval_crit = nn.L1Loss()
+# adamw 
+optimizer = optim.AdamW(model.parameters(), lr=0.001, weight_decay=0.001)
 best_loss = np.inf  # Initialize the best loss to infinity
 start_epoch = 0
+scheduler = optim.lr_scheduler.StepLR(optimizer, step_size=10, gamma=0.5)
 
 
 predicted_ages = None
 # Training loop
-epochs = 30
+epochs = int(sys.argv[3])
+
+# Initialize lists to track loss
+filename = sys.argv[1] 
+csv_file = f"model_dumps/mix/{slice_count}/{filename}.csv"
+
 for epoch in range(start_epoch, epochs):
     model.train()
     train_loss = 0.0
     predicted_ages = []
+
     for idx, (features, sex, age) in enumerate(train_loader):
         features = features.unsqueeze(1).to(device)  # Add channel dimension
         sex = sex.to(device)
         age = age.to(device)
         optimizer.zero_grad()
         outputs = model(features, sex)
-        train_outputs[train_indices[idx]] = outputs.item()
-        # print (outputs)
+        # Store the output for each sample in the batch
+        for i in range(outputs.size(0)):
+            train_outputs[train_indices[idx * batch_size + i]] = outputs[i].item()
+
         loss = criterion(outputs.squeeze(), age)
-        
         loss.backward()
         optimizer.step()
 
@@ -74,7 +449,6 @@
     # Validation loop
     model.eval()
     val_loss = 0.0
-    # predicted_ages = []
     with torch.no_grad():
         for idx, (features, sex, age) in enumerate(val_loader):
             features = features.unsqueeze(1).to(device)
@@ -82,42 +456,20 @@
             age = age.to(device)
 
             outputs = model(features, sex)
-            # predicted_ages.append(outputs.item())
-            loss = criterion(outputs.squeeze(), age)
-
+            loss = eval_crit(outputs.squeeze(), age)
             val_loss += loss.item()
 
-             # Save the predicted age for the current validation sample
-            val_outputs[val_indices[idx]] = outputs.item()
+            # Save the predicted age for the current validation sample
+            for i in range(outputs.size(0)):
+                val_outputs[val_indices[idx * batch_size + i]] = outputs[i].item()
+            # val_outputs[val_indices[idx]] = outputs.item()
 
-    # print (predicted_ages)
     val_loss /= len(val_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Validation Loss: {val_loss:.4f}")
+
 
- 
-    max_index = max(train_outputs.keys())
-    # Create a DataFrame with NaN for all indices initially
-    df_trn = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
-    # Assign the values to their respective indices
-    for index, value in train_outputs.items():
-        df_trn.loc[index, "Predicted_Age"] = value
-    print (df_trn)
-
-    df2 = df.copy()
-    df2['Predicted_Age'] = df_trn['Predicted_Age']
-    train_df = df2.loc[train_outputs.keys()]
-    print (train_df)
-    train_df.to_csv("predicted_ages_train.csv")
-
-    max_index = max(val_outputs.keys())
-    # Create a DataFrame with NaN for all indices initially
-    df_pred = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
-    # Assign the values to their respective indices
-    for index, value in val_outputs.items():
-        df_pred.loc[index, "Predicted_Age"] = value
-    print (df_pred)
-
-    df1 = df.copy()
-    df1['Predicted_Age'] = df_pred['Predicted_Age']
-    test_df = df1.loc[val_outputs.keys()]
-    print (test_df)
-    test_df.to_csv("predicted_ages_val.csv")
+    # Check if validation loss improved
+    if val_loss < best_loss:
+        best_loss = val_loss
+        print(f"Validation loss improved to {best_loss:.4f}. Saving model...")
+       
diff --git a/dataset_cls.py b/dataset_cls.py
new file mode 100644
index 0000000..6971413
--- /dev/null
+++ b/dataset_cls.py
@@ -0,0 +1,68 @@
+# Custom Dataset
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, Dataset
+class ADNIDataset(Dataset):
+    def __init__(self, features_list, sex_list, age_list):
+        self.features = features_list
+        self.sex = sex_list
+        self.age = age_list
+
+    def __len__(self):
+        return len(self.age)
+
+    def __getitem__(self, idx):
+        return (
+            torch.tensor(self.features[idx], dtype=torch.float32),
+            torch.tensor(self.sex[idx], dtype=torch.float32),
+            torch.tensor(self.age[idx], dtype=torch.float32),
+        )
+    
+
+class ADNIDatasetV2(Dataset):
+    def __init__(self, features_list, sex_list, age_list):
+        self.features = features_list
+        self.sex = sex_list
+        self.age = age_list
+
+    def __len__(self):
+        return len(self.age)
+
+    def __getitem__(self, idx):
+        return (
+            torch.tensor(self.features[idx], dtype=torch.float32),
+            torch.tensor(self.sex[idx], dtype=torch.float32),
+            self.age[idx]
+        )
+
+
+class ADNIDatasetLite(Dataset):
+    def __init__(self,image_list, sex_list, age_list):
+        self.images = image_list
+        self.sex = sex_list
+        self.age = age_list
+
+    def __len__(self):
+        return len(self.age)
+
+    def __getitem__(self, idx):
+        return (
+            torch.tensor(self.images[idx], dtype=torch.float32),
+            torch.tensor(self.sex[idx], dtype=torch.float32),
+            torch.tensor(self.age[idx], dtype=torch.float32),
+        )
+
+class ADNIDatasetViT(Dataset):
+    def __init__(self,image_list, age_list):
+        self.images = image_list
+        self.label = age_list
+
+    def __len__(self):
+        return len(self.label)
+
+    def __getitem__(self, idx):
+        return (
+            self.images[idx],
+            self.label[idx],
+        )
diff --git a/dataviz.py b/dataviz.py
new file mode 100644
index 0000000..b0cd3ea
--- /dev/null
+++ b/dataviz.py
@@ -0,0 +1,232 @@
+import os
+import pandas as pd
+import nibabel as nib
+import numpy as np
+import torch
+import torch.nn as nn
+import torch.nn.init as init
+import torch.optim as optim
+from tqdm import tqdm
+from nibabel.orientations import io_orientation, axcodes2ornt, ornt_transform, apply_orientation
+from torchvision import transforms
+from transformers import ViTFeatureExtractor, ViTModel
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler
+import pickle
+import matplotlib.pyplot as plt
+import sys
+import SimpleITK as sitk
+from scipy.ndimage import zoom
+from dataset_cls import ADNIDatasetV2, ADNIDatasetViT
+from torch.utils.data import DataLoader, Dataset
+import gc
+
+# def set_random_seed(seed=69420):
+#     np.random.seed(seed)
+#     torch.manual_seed(seed)
+#     torch.cuda.manual_seed(seed)
+#     torch.cuda.manual_seed_all(seed)  # For multi-GPU setups
+#     torch.backends.cudnn.deterministic = True
+#     torch.backends.cudnn.benchmark = False
+
+# universal_seed = 69420
+
+# set_random_seed(universal_seed)
+
+# # Function to resample the volume to 160 slices
+# def resample_volume_to_fixed_slices(data, affine, target_slices=160):
+#     # Convert Numpy array and affine to SimpleITK image
+#     sitk_img = sitk.GetImageFromArray(data)
+#     sitk_img.SetSpacing([affine[0, 0], affine[1, 1], affine[2, 2]])
+
+#     original_size = sitk_img.GetSize()  # (width, height, depth)
+#     original_spacing = sitk_img.GetSpacing()  # (spacing_x, spacing_y, spacing_z)
+
+#     # Calculate new spacing to achieve the target number of slices
+#     new_spacing = list(original_spacing)
+#     new_spacing[2] = (original_spacing[2] * original_size[2]) / target_slices
+
+#     # Define new size
+#     new_size = [original_size[0], original_size[1], target_slices]
+
+#     # Resample the image
+#     resampler = sitk.ResampleImageFilter()
+#     resampler.SetOutputSpacing(new_spacing)
+#     resampler.SetSize(new_size)
+#     resampler.SetInterpolator(sitk.sitkLinear)
+#     resampled_img = resampler.Execute(sitk_img)
+
+#     return sitk.GetArrayFromImage(resampled_img)  # Return the resampled image as a numpy array
+
+
+# def resample_nifti(img_data, target_slices = 160):
+#     # Determine the current number of slices along the z-axis (3rd dimension)
+#     current_slices = img_data.shape[0]
+#     # Calculate the zoom factor for resampling (only along the z-axis)
+#     zoom_factor = target_slices / current_slices
+#     # Resample the image data along the z-axis
+#     resampled_data = zoom(img_data, (zoom_factor, 1, 1), order=3)  # order=3 for cubic interpolation
+#     # Ensure that the resampled data has the target number of slices
+#     # print (resampled_data.shape)
+#     # resampled_data = resampled_data[:target_slices,:,:]
+#     # print (resampled_data.shape)
+#     return resampled_data
+
+
+# # Check if GPU is available
+# device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+# # Load the CSV file into a pandas DataFrame
+# csv_path = "adni_storage/adni_brainrotnet_metadata.csv"
+# df_adni = pd.read_csv(csv_path)
+# # df = df.sample(n=1000, random_state=69420)
+# # Add a new column 'filepath' with the constructed file paths
+# df_adni['filepath'] = df_adni.apply(
+#     lambda row: f"adni_storage/ADNI_nii_gz_bias_corrected/I{row['ImageID'][4:]}_{row['SubjectID']}.stripped.N4.nii.gz",
+#     axis=1
+# )
+# df_adni = df_adni.sort_values(by='Age', ascending=True).reset_index(drop=True).head(500)
+
+
+# metadata_path = "dlbs_storage/dlbs_brainrotnet_metadata.csv"
+# df_dlbs = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_dlbs['filepath'] = df_dlbs.apply(
+#     lambda row: f"dlbs_storage/DLBS_bias_corrected/{row['ImageID'][4:]}.stripped.N4.nii.gz",
+#     axis=1
+# )
+
+
+# metadata_path = "fcon1000_storage/fcon1000_brainrotnet_metadata.csv"
+# df_fcon = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_fcon['filepath'] = df_fcon.apply(
+#     lambda row: f"fcon1000_storage/fcon1000_bias_corrected/{row['ImageID'][8:]}.stripped.N4.nii.gz",
+#     axis=1
+# )
+# df_fcon = df_fcon.dropna()
+
+# metadata_path = "camcan_storage/camcan_brainrotnet_metadata.csv"
+# df_camcan = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_camcan['filepath'] = df_camcan.apply(
+#     lambda row: f"camcan_storage/CamCAN_nii_gz_bias_corrected/{row['ImageID']}.stripped.N4.nii.gz",
+#     axis=1
+# )
+
+# df = pd.concat([
+#     df_adni[['ImageID', 'Sex', 'Age', 'filepath']], 
+#     df_dlbs[['ImageID', 'Sex', 'Age', 'filepath']],
+#     df_fcon[['ImageID', 'Sex', 'Age', 'filepath']],
+#     df_camcan[['ImageID', 'Sex', 'Age', 'filepath']]
+# ], ignore_index=True)
+
+# class DataFrameDataset(Dataset):
+#     def __init__(self, dataframe):
+#         self.dataframe = dataframe
+
+#     def __len__(self):
+#         return len(self.dataframe)
+
+#     def __getitem__(self, idx):
+#         return self.dataframe.iloc[idx]
+
+# wrapped_dataset = DataFrameDataset(df)
+
+# generator = torch.Generator().manual_seed(universal_seed)
+
+# train_size = int(0.8 * len(wrapped_dataset))
+# val_size = len(wrapped_dataset) - train_size
+# train_dataset, val_dataset = torch.utils.data.random_split(wrapped_dataset, [train_size, val_size], generator=generator)
+
+# df_train = pd.DataFrame([item for item in train_dataset])
+# df_val = pd.DataFrame([item for item in val_dataset])
+
+# df_train.to_csv("train_dataset.csv", index=False)
+# df_val.to_csv("val_dataset.csv", index=False)
+
+# Load datasets
+# df_train = pd.read_csv("train_dataset.csv")
+# df_val = pd.read_csv("val_dataset.csv")
+
+#combine the datasets and save them in a csv
+
+# df_combined = pd.concat([df_train, df_val], ignore_index=True)
+# df_combined.to_csv("combined_dataset.csv", index=False)
+# Load the combined dataset
+
+
+# import pandas as pd
+# import numpy as np
+
+# df = pd.read_csv("combined_dataset.csv")
+
+# # Initialize dictionaries to store age data for each dataset
+# adni_ages = []
+# dlbs_ages = []
+# fcon_ages = []
+# camcan_ages = []
+
+# # Loop through the dataframe and classify rows based on file path
+# for index, row in df.iterrows():
+#     if 'adni_storage' in row['filepath']:
+#         adni_ages.append(row['Age'])
+#     elif 'dlbs_storage' in row['filepath']:
+#         dlbs_ages.append(row['Age'])
+#     elif 'fcon1000_storage' in row['filepath']:
+#         fcon_ages.append(row['Age'])
+#     elif 'camcan_storage' in row['filepath']:
+#         camcan_ages.append(row['Age'])
+
+# # TOTAL NUMBER OF IMAGES
+# print(f"ADNI: {len(adni_ages)}")
+# print(f"DLBS: {len(dlbs_ages)}")
+# print(f"FCON: {len(fcon_ages)}")
+# print(f"Cam-CAN: {len(camcan_ages)}")
+
+# # Calculate mean and standard deviation for each dataset
+# adni_mean, adni_std = np.mean(adni_ages), np.std(adni_ages)
+# dlbs_mean, dlbs_std = np.mean(dlbs_ages), np.std(dlbs_ages)
+# fcon_mean, fcon_std = np.mean(fcon_ages), np.std(fcon_ages)
+# camcan_mean, camcan_std = np.mean(camcan_ages), np.std(camcan_ages)
+
+# # Print the results
+# print(f"ADNI: {adni_mean:.2f} ± {adni_std:.2f}")
+# print(f"DLBS: {dlbs_mean:.2f} ± {dlbs_std:.2f}")
+# print(f"FCON: {fcon_mean:.2f} ± {fcon_std:.2f}")
+# print(f"Cam-CAN: {camcan_mean:.2f} ± {camcan_std:.2f}")
+
+# # compute the range of ages for each dataset
+# print("Age ranges:")
+# print(f"ADNI: {min(adni_ages)} - {max(adni_ages)}")
+# print(f"DLBS: {min(dlbs_ages)} - {max(dlbs_ages)}")
+# print(f"FCON: {min(fcon_ages)} - {max(fcon_ages)}")
+# print(f"Cam-CAN: {min(camcan_ages)} - {max(camcan_ages)}")
+
+# # Load the datasets
+
+# df_train = pd.read_csv("train_dataset.csv")
+# df_val = pd.read_csv("val_dataset.csv")
+
+# print(len(df_train))
+# print(len(df_val))
+
+# count_train_adni = df_train['filepath'].str.contains('adni_storage').sum()
+# count_train_dlbs = df_train['filepath'].str.contains('dlbs_storage').sum()
+# count_train_fcon = df_train['filepath'].str.contains('fcon1000_storage').sum()
+# count_train_camcan = df_train['filepath'].str.contains('camcan_storage').sum()
+# print(f"Train ADNI: {count_train_adni}")
+# print(f"Train DLBS: {count_train_dlbs}")
+# print(f"Train FCON: {count_train_fcon}")
+# print(f"Train Cam-CAN: {count_train_camcan}")
+
+# count_val_adni = df_val['filepath'].str.contains('adni_storage').sum()
+# count_val_dlbs = df_val['filepath'].str.contains('dlbs_storage').sum()
+# count_val_fcon = df_val['filepath'].str.contains('fcon1000_storage').sum()
+# count_val_camcan = df_val['filepath'].str.contains('camcan_storage').sum()
+
+# print(f"Val ADNI: {count_val_adni}")
+# print(f"Val DLBS: {count_val_dlbs}")
+# print(f"Val FCON: {count_val_fcon}")
+# print(f"Val Cam-CAN: {count_val_camcan}")
+
diff --git a/db3_adni.py b/db3_adni.py
index ecd0040..90cd888 100644
--- a/db3_adni.py
+++ b/db3_adni.py
@@ -19,8 +19,8 @@
 
 
 # Define the input and output directories
-input_dir = "adni_storage/ADNI_nii_gz"
-output_dir = "adni_storage/ADNI_nii_gz_stripped"
+input_dir = "camcan_storage/CamCAN"
+output_dir = "camcan_storage/CamCAN_nii_gz_stripped"
 
 # Ensure the output directory exists
 os.makedirs(output_dir, exist_ok=True)
diff --git a/deit_train_feature_cnn_main_mix_roi.py b/deit_train_feature_cnn_main_mix_roi.py
new file mode 100644
index 0000000..4c31a0b
--- /dev/null
+++ b/deit_train_feature_cnn_main_mix_roi.py
@@ -0,0 +1,1309 @@
+import os
+import pandas as pd
+import nibabel as nib
+import numpy as np
+import torch
+import torch.nn as nn
+import torch.nn.init as init
+import torch.optim as optim
+from tqdm import tqdm
+from nibabel.orientations import io_orientation, axcodes2ornt, ornt_transform, apply_orientation
+from torchvision import transforms
+from transformers import ViTFeatureExtractor, ViTModel
+from transformers import ViTForImageClassification
+from transformers import DeiTForImageClassification
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler
+import pickle
+import matplotlib.pyplot as plt
+import sys
+import SimpleITK as sitk
+from scipy.ndimage import zoom
+from dataset_cls import ADNIDataset, ADNIDatasetViT
+from torch.utils.data import DataLoader, Dataset
+import gc
+
+model_name = "deit"
+
+def set_random_seed(seed=69420):
+    np.random.seed(seed)
+    torch.manual_seed(seed)
+    torch.cuda.manual_seed(seed)
+    torch.cuda.manual_seed_all(seed)  # For multi-GPU setups
+    torch.backends.cudnn.deterministic = True
+    torch.backends.cudnn.benchmark = False
+
+universal_seed = 69420
+
+set_random_seed(universal_seed)
+
+# Function to resample the volume to 160 slices
+def resample_volume_to_fixed_slices(data, affine, target_slices=160):
+    # Convert Numpy array and affine to SimpleITK image
+    sitk_img = sitk.GetImageFromArray(data)
+    sitk_img.SetSpacing([affine[0, 0], affine[1, 1], affine[2, 2]])
+
+    original_size = sitk_img.GetSize()  # (width, height, depth)
+    original_spacing = sitk_img.GetSpacing()  # (spacing_x, spacing_y, spacing_z)
+
+    # Calculate new spacing to achieve the target number of slices
+    new_spacing = list(original_spacing)
+    new_spacing[2] = (original_spacing[2] * original_size[2]) / target_slices
+
+    # Define new size
+    new_size = [original_size[0], original_size[1], target_slices]
+
+    # Resample the image
+    resampler = sitk.ResampleImageFilter()
+    resampler.SetOutputSpacing(new_spacing)
+    resampler.SetSize(new_size)
+    resampler.SetInterpolator(sitk.sitkLinear)
+    resampled_img = resampler.Execute(sitk_img)
+
+    return sitk.GetArrayFromImage(resampled_img)  # Return the resampled image as a numpy array
+
+
+def resample_nifti(img_data, target_slices = 160):
+    # Determine the current number of slices along the z-axis (3rd dimension)
+    current_slices = img_data.shape[0]
+    # Calculate the zoom factor for resampling (only along the z-axis)
+    zoom_factor = target_slices / current_slices
+    # Resample the image data along the z-axis
+    resampled_data = zoom(img_data, (zoom_factor, 1, 1), order=3)  # order=3 for cubic interpolation
+    # Ensure that the resampled data has the target number of slices
+    # print (resampled_data.shape)
+    # resampled_data = resampled_data[:target_slices,:,:]
+    # print (resampled_data.shape)
+    return resampled_data
+
+
+# Check if GPU is available
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+# Load the CSV file into a pandas DataFrame
+csv_path = "adni_storage/adni_brainrotnet_metadata.csv"
+df_adni = pd.read_csv(csv_path)
+# df = df.sample(n=1000, random_state=69420)
+# Add a new column 'filepath' with the constructed file paths
+df_adni['filepath'] = df_adni.apply(
+    lambda row: f"adni_storage/ADNI_nii_gz_bias_corrected/I{row['ImageID'][4:]}_{row['SubjectID']}.stripped.N4.nii.gz",
+    axis=1
+)
+df_adni = df_adni.sort_values(by='Age', ascending=True).reset_index(drop=True).head(500)
+# df_adni=df_adni.sample(n=400)
+
+# Load independent dataset metadata
+# metadata_path = "ixi_storage/ixi_brainrotnet_metadata.csv"
+# df_ixi = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_ixi['filepath'] = df_ixi.apply(
+#     lambda row: f"ixi_storage/IXI_bias_corrected/{row['ImageID']}.stripped.N4.nii.gz",
+#     axis=1
+# )
+
+# metadata_path = "abide_storage/abide_brainrotnet_metadata.csv"
+# df_abide = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_abide['filepath'] = df_abide.apply(
+#     lambda row: f"abide_storage/ABIDEII_bias_corrected/{row['ImageID'][7:]}.stripped.N4.nii.gz",
+#     axis=1
+# )
+# df_abide = df_abide.sort_values(by='Age', ascending=False).reset_index(drop=True).head(500)
+# df_abide=df_abide.sample(n=400)
+
+metadata_path = "dlbs_storage/dlbs_brainrotnet_metadata.csv"
+df_dlbs = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_dlbs['filepath'] = df_dlbs.apply(
+    lambda row: f"dlbs_storage/DLBS_bias_corrected/{row['ImageID'][4:]}.stripped.N4.nii.gz",
+    axis=1
+)
+
+# metadata_path = "cobre_storage/cobre_brainrotnet_metadata.csv"
+# df_cobre = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_cobre['filepath'] = df_cobre.apply(
+#     lambda row: f"cobre_storage/COBRE_bias_corrected/{row['ImageID'][5:]}.stripped.N4.nii.gz",
+#     axis=1
+# )
+
+metadata_path = "fcon1000_storage/fcon1000_brainrotnet_metadata.csv"
+df_fcon = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_fcon['filepath'] = df_fcon.apply(
+    lambda row: f"fcon1000_storage/fcon1000_bias_corrected/{row['ImageID'][8:]}.stripped.N4.nii.gz",
+    axis=1
+)
+df_fcon = df_fcon.dropna()
+# df_fcon = df_fcon.sort_values(by='Age', ascending=False).reset_index(drop=True).head(300)
+# df_fcon = df_fcon.sample(n=300)
+
+metadata_path = "sald_storage/sald_brainrotnet_metadata.csv"
+df_sald = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_sald['filepath'] = df_sald.apply(
+    lambda row: f"sald_storage/SALD_bias_corrected/sub-{row['ImageID'][4:]}.stripped.N4.nii.gz",
+    axis=1
+)
+df_sald = df_sald.sort_values(by='Age', ascending=False).reset_index(drop=True).head(300)
+# df_sald = df_sald.sample(n=300)
+
+# metadata_path = "corr_storage/corr_brainrotnet_metadata.csv"
+# df_corr = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_corr['filepath'] = df_corr.apply(
+#     lambda row: f"corr_storage/CORR_bias_corrected/{row['ImageID'][5:]}.stripped.N4.nii.gz",
+#     axis=1
+# )
+# df_corr = df_corr.sort_values(by='Age', ascending=True).reset_index(drop=True).head(300)
+# df_corr = df_corr.sample(n=200)
+
+
+metadata_path = "oasis1_storage/oasis1_brainrotnet_metadata.csv"
+df_oas1 = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_oas1['filepath'] = df_oas1.apply(
+    lambda row: f"oasis1_storage/oasis_nii_gz_bias_corrected/{row['ImageID']}.stripped.N4.nii.gz",
+    axis=1
+)
+df_oas1 = df_oas1.sort_values(by='Age', ascending=False).reset_index(drop=True).head(300)
+metadata_path = "camcan_storage/camcan_brainrotnet_metadata.csv"
+df_camcan = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_camcan['filepath'] = df_camcan.apply(
+    lambda row: f"camcan_storage/CamCAN_nii_gz_bias_corrected/{row['ImageID']}.stripped.N4.nii.gz",
+    axis=1
+)
+
+
+df = pd.concat ([
+                 df_adni[['ImageID', 'Sex', 'Age', 'filepath']], 
+                #  df_ixi[['ImageID', 'Sex', 'Age', 'filepath']], 
+                #  df_abide[['ImageID', 'Sex', 'Age', 'filepath']],
+                 df_dlbs[['ImageID', 'Sex', 'Age', 'filepath']],
+                #  df_cobre[['ImageID', 'Sex', 'Age', 'filepath']],
+                 df_fcon[['ImageID', 'Sex', 'Age', 'filepath']],
+                 df_sald[['ImageID', 'Sex', 'Age', 'filepath']],
+                #  df_corr[['ImageID', 'Sex', 'Age', 'filepath']], 
+                 df_oas1[['ImageID', 'Sex', 'Age', 'filepath']],
+                df_camcan[['ImageID', 'Sex', 'Age', 'filepath']]
+                 ], ignore_index=True)
+print (df)
+# Ensure 'Age' is an integer
+df['Age_Group'] = df['Age'].astype(int).apply(lambda x: f"{x:03d}"[:-1] + "0")
+df['Age_Group'] = df['Age_Group'] + df['Sex']
+print (df['Age_Group'].unique())
+# Prepare dataset and dataloaders
+sex_encoded = df['Sex'].apply(lambda x: 0 if x == 'M' else 1).tolist()
+age_list = df['Age'].tolist()
+filepath_list = df['filepath'].tolist()
+label_list = df['Age_Group'].tolist()
+
+# Get unique labels and create a mapping
+unique_labels = sorted(set(label_list))  # Ensure consistent ordering
+label_to_idx = {label: idx for idx, label in enumerate(unique_labels)}
+idx_to_label = {idx: label for label, idx in label_to_idx.items()}  # Reverse mapping for decoding
+
+# Convert labels to integers
+numeric_labels = [label_to_idx[label] for label in label_list]
+label_list = numeric_labels
+
+roi = 160
+
+# Transformation pipeline for ViT
+transform = transforms.Compose([
+    transforms.ToPILImage(),
+    transforms.Resize((224, 224)),
+    transforms.Lambda(lambda img: img.convert("RGB")),  # Convert to RGB
+    transforms.ToTensor(),
+    transforms.Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5]),  # Normalize for ViT
+])
+
+
+# Function to extract 16 evenly spaced slices
+def extract_slices(volume, num_slices=16):
+    total_slices = volume.shape[0]
+    indices = np.linspace(0, total_slices - 1, num_slices, dtype=int)
+    return volume[indices, :, :]  # Select slices
+
+def calculate_bounding_box_from_volume(volume):
+    # Find indices of non-zero values
+    non_zero_indices = np.argwhere(volume > 0)
+
+    # Calculate min and max indices along each dimension
+    min_indices = np.min(non_zero_indices, axis=0)
+    max_indices = np.max(non_zero_indices, axis=0)
+
+    # Convert indices to integers
+    min_indices = min_indices.astype(int)
+    max_indices = max_indices.astype(int)
+
+    return min_indices, max_indices
+
+
+def crop_brain_volumes(brain_data):
+    
+
+        # Calculate bounding box from the brain volume
+    min_indices, max_indices = calculate_bounding_box_from_volume(brain_data)
+
+        # Crop the volume
+    cropped_brain = brain_data[min_indices[0]:max_indices[0] + 1,
+                                   min_indices[1]:max_indices[1] + 1,
+                                   min_indices[2]:max_indices[2] + 1]
+    return cropped_brain
+
+# Function to preprocess data and dynamically expand slices while saving to disk
+def preprocess_and_expand(dataset, transform, output_dir, num_slices=16):
+    os.makedirs(output_dir, exist_ok=True)  # Ensure output directory exists
+    expanded_images, expanded_labels = [], []
+
+    for filepath, label in tqdm(dataset, desc="Processing Slices"):
+        # Check if all slice files already exist
+        all_slices_exist = True
+        slice_filenames = [
+            os.path.join(output_dir, f"{os.path.basename(filepath)}_slice_{i}.pt")
+            for i in range(num_slices)
+        ]
+        if not all(os.path.exists(slice_file) for slice_file in slice_filenames):
+            all_slices_exist = False
+
+        # Skip processing if all slices exist
+        if all_slices_exist:
+            expanded_images.extend(slice_filenames)  # Add existing file paths
+            expanded_labels.extend([label] * num_slices)
+            continue
+
+        # Load NIfTI image only if slices are missing
+        nii_img = nib.load(filepath)
+        orig_ornt = io_orientation(nii_img.affine)
+        ras_ornt = axcodes2ornt(("R", "A", "S"))
+        ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+        data = nii_img.get_fdata()  # Load image data
+        data = apply_orientation(data, ornt_trans)
+
+        data = crop_brain_volumes(data)
+
+        # Normalize and extract slices
+        data = (data - data.min()) / (data.max() - data.min())
+        slices = extract_slices(data, num_slices)
+
+        # Transform each slice, save to file, and add to dataset
+        for i, slice_data in enumerate(slices):
+            slice_filename = slice_filenames[i]
+            if not os.path.exists(slice_filename):
+                transformed_slice = transform(slice_data)  # Transform slice
+                torch.save(transformed_slice, slice_filename)  # Save to file
+            expanded_images.append(slice_filename)  # Store file path
+            expanded_labels.append(label)
+
+    return expanded_images, expanded_labels
+# Instantiate Dataset
+vit_dataset = ADNIDatasetViT(filepath_list, label_list)
+
+# Split Dataset
+train_size = int(0.8 * len(vit_dataset))
+val_size = len(vit_dataset) - train_size
+generator = torch.Generator().manual_seed(universal_seed)
+vit_train_dataset, vit_val_dataset = torch.utils.data.random_split(vit_dataset, [train_size, val_size], generator=generator)
+
+# Create New Dataset with Filepaths
+class ExpandedDataset(Dataset):
+    def __init__(self, image_paths, labels):
+        self.image_paths = image_paths
+        self.labels = labels
+
+    def __len__(self):
+        return len(self.labels)
+
+    def __getitem__(self, idx):
+        # Load the image from file
+        image = torch.load(self.image_paths[idx])
+        label = torch.tensor(self.labels[idx], dtype=torch.long)
+        return image, label
+
+slice_count = 32
+
+# Define output directory for slices
+output_dir = f"processed_slices/{slice_count}"
+
+# Preprocess and expand the training data
+expanded_image_paths, expanded_labels = preprocess_and_expand(vit_train_dataset, transform, output_dir, num_slices=slice_count)
+
+# Create Expanded Dataset and DataLoader
+expanded_train_dataset = ExpandedDataset(expanded_image_paths, expanded_labels)
+expanded_train_loader = DataLoader(expanded_train_dataset, batch_size=8, shuffle=True)
+
+# Print Sizes
+print(f"Original Training Dataset Size: {len(vit_train_dataset)}")
+print(f"Expanded Training Dataset Size: {len(expanded_train_dataset)}")
+
+
+# Load ViT model
+num_classes = df['Age_Group'].nunique()  # Number of unique Age_Groups
+# model = ViTForImageClassification.from_pretrained(
+#     "google/vit-base-patch16-224",
+#     num_labels=num_classes,
+#     ignore_mismatched_sizes=True, 
+# )
+
+model = DeiTForImageClassification.from_pretrained(
+    "facebook/deit-base-distilled-patch16-224",
+    num_labels=num_classes,
+    ignore_mismatched_sizes=True,
+)
+
+
+model.to(device)
+
+# Loss function and optimizer
+criterion = nn.CrossEntropyLoss()
+optimizer = optim.AdamW(model.parameters(), lr=1e-4)
+
+# Function to save checkpoint
+def save_checkpoint(epoch, model, optimizer, path=f"model_dumps/{model_name}_train_checkpoint_{slice_count}.pth"):
+    torch.save({
+        'epoch': epoch,
+        'model_state_dict': model.state_dict(),
+        'optimizer_state_dict': optimizer.state_dict(),
+    }, path)
+    print(f"Checkpoint saved at epoch {epoch+1}")
+
+# Function to load checkpoint
+def load_checkpoint(path=f"model_dumps/{model_name}_train_checkpoint_{slice_count}.pth"):
+    checkpoint = torch.load(path)
+    model.load_state_dict(checkpoint['model_state_dict'])
+    optimizer.load_state_dict(checkpoint['optimizer_state_dict'])
+    start_epoch = checkpoint['epoch'] + 1
+    print(f"Checkpoint loaded. Resuming from epoch {start_epoch}")
+    return start_epoch
+
+# Check if recovery mode is enabled
+checkpoint_path = f"model_dumps/{model_name}_train_checkpoint_{slice_count}.pth"
+start_epoch = 0
+
+if len(sys.argv) > 4 and sys.argv[4] == "recover":
+    start_epoch = load_checkpoint(path=checkpoint_path)
+
+# Training loop
+vit_train_epochs = 5
+model.train()
+
+for epoch in range(start_epoch, vit_train_epochs):
+    running_loss = 0.0
+    correct = 0
+    total = 0
+
+    for inputs, labels in tqdm(expanded_train_loader, desc=f"Epoch {epoch+1}/{vit_train_epochs}"):
+        # Move data to device
+        inputs, labels = inputs.to(device), labels.to(device)
+
+        # Forward pass
+        outputs = model(pixel_values=inputs)  # ViT expects `pixel_values`
+        loss = criterion(outputs.logits, labels)
+
+        # Backward pass and optimization
+        optimizer.zero_grad()
+        loss.backward()
+        optimizer.step()
+
+        running_loss += loss.item()
+
+        # Calculate accuracy
+        _, predicted = torch.max(outputs.logits, dim=1)
+        total += labels.size(0)
+        correct += (predicted == labels).sum().item()
+
+        # batch_accuracy = 100 * (predicted == labels).sum().item() / labels.size(0)
+        # print(f"Batch Accuracy: {batch_accuracy:.2f}%")
+
+    epoch_loss = running_loss / len(expanded_train_loader)
+    epoch_accuracy = 100 * correct / total
+    print(f"Epoch {epoch+1}/{vit_train_epochs}, Loss: {epoch_loss:.4f}, Accuracy: {epoch_accuracy:.2f}%")
+    # sav the accuracy and loss for each epoch in a csv file
+    with open(f'{model_name}_train_metrics_{slice_count}.csv', 'a') as f:
+        f.write(f"{epoch+1},{epoch_loss},{epoch_accuracy}\n")
+    # Save checkpoint at the end of each epoch
+    # shift the model to cpu before saving
+    # model.to('cpu')
+    save_checkpoint(epoch, model, optimizer, path=f"model_dumps/{model_name}_train_checkpoint_{slice_count}.pth")
+    # model.to(device)  # Move back to GPU
+    gc.collect()
+
+# Load pre-trained ViT model
+feature_extractor = ViTFeatureExtractor.from_pretrained("google/vit-base-patch16-224")
+# model = ViTModel.from_pretrained("google/vit-base-patch16-224")
+# model.to(device)  # Move the model to the GPU (if available)
+model.eval()
+
+# Update image transform for grayscale images to match ViT input requirements
+transform = transforms.Compose([
+    transforms.ToPILImage(),
+    transforms.Resize((224, 224)),
+    transforms.Lambda(lambda img: img.convert("RGB")),  # Convert to RGB directly
+    transforms.ToTensor(),
+    transforms.Normalize(mean=feature_extractor.image_mean, std=feature_extractor.image_std),
+])
+
+torch.cuda.empty_cache()  # Free GPU memory
+
+
+
+# To store features and labels
+features_list = []
+labels_list = []
+
+
+# Directory to save processed images and features
+os.makedirs(f"adni_storage/ADNI_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+# Process each row in the DataFrame,
+for _, row in tqdm(df_adni.iterrows(), total=len(df_adni), desc="Processing images"):
+    filepath = row['filepath']
+    image_title = f"{row['ImageID'][4:]}_{row['SubjectID']}"
+
+    # Check if the feature file already exists
+    feature_file_path = f"adni_storage/ADNI_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        
+        from PIL import Image
+        # Normalize the array to 0-255 for grayscale image
+        data_normalized = ((features - np.min(features)) / (np.max(features) - np.min(features)) * 255).astype(np.uint8)
+        data_normalized = np.repeat(data_normalized, 4, axis=0)
+        # Create an image from the array
+        img = Image.fromarray(np.transpose(data_normalized), mode='L')  # 'L' mode for grayscale
+        # Save the image
+        # img.save(f"adni_storage/ADNI_features/train_e{vit_train_epochs}/{slice_count}/featuremaps/{image_title}_fm.png")
+
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        # print ("hiii")
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                data = crop_brain_volumes(data)
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+
+                # Extract features for all sagittal slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+
+                    # Transform slice for ViT input
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)  # Add batch dimension and move to GPU
+
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        # #outputs = model(slice_tensor)
+                        # slice_features = model.deit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()  # Move output back to CPU
+                        slice_features = model.deit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Target is 'Age'
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+
+# Directory to save processed images and features
+# os.makedirs(f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+# for _, row in tqdm(df_ixi.iterrows(), total=len(df_ixi), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID']}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         #outputs = model(slice_tensor)
+#                         slice_features = model.deit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+
+# # Directory to save processed images and features
+# os.makedirs(f"abide_storage/ABIDEII_features/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+# for _, row in tqdm(df_abide.iterrows(), total=len(df_abide), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][7:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"abide_storage/ABIDEII_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         #outputs = model(slice_tensor)
+#                         slice_features = model.deit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+os.makedirs(f"dlbs_storage/DLBS_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+for _, row in tqdm(df_dlbs.iterrows(), total=len(df_dlbs), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID'][4:]}"
+        # Check if the feature file already exists
+    feature_file_path = f"dlbs_storage/DLBS_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+    # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                data = crop_brain_volumes(data)
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        #outputs = model(slice_tensor)
+                        slice_features = model.deit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+
+# os.makedirs(f"cobre_storage/COBRE_features/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+# for _, row in tqdm(df_cobre.iterrows(), total=len(df_cobre), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][5:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"cobre_storage/COBRE_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         #outputs = model(slice_tensor)
+#                         slice_features = model.deit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+os.makedirs(f"fcon1000_storage/fcon1000_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+for _, row in tqdm(df_fcon.iterrows(), total=len(df_fcon), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID'][5:]}"
+        # Check if the feature file already exists
+    feature_file_path = f"fcon1000_storage/fcon1000_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+    # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                data = crop_brain_volumes(data)
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        #outputs = model(slice_tensor)
+                        slice_features = model.deit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+os.makedirs(f"sald_storage/SALD_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+for _, row in tqdm(df_sald.iterrows(), total=len(df_sald), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID'][4:]}"
+        # Check if the feature file already exists
+    feature_file_path = f"sald_storage/SALD_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+    # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        #outputs = model(slice_tensor)
+                        slice_features = model.deit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+# os.makedirs(f"corr_storage/CORR_features/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+# for _, row in tqdm(df_corr.iterrows(), total=len(df_corr), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][5:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"corr_storage/CORR_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         #outputs = model(slice_tensor)
+#                         slice_features = model.deit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+
+os.makedirs(f"oasis1_storage/oasis1_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+for _, row in tqdm(df_oas1.iterrows(), total=len(df_oas1), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID']}"
+        # Check if the feature file already exists
+    feature_file_path = f"oasis1_storage/oasis1_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        #outputs = model(slice_tensor)
+                        slice_features = model.deit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+os.makedirs(f"camcan_storage/CamCAN_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+for _, row in tqdm(df_camcan.iterrows(), total=len(df_camcan), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID']}"
+        # Check if the feature file already exists
+    feature_file_path = f"camcan_storage/CamCAN_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+    # feature_file_path = f"ixi_storage/IXI_features/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                data = crop_brain_volumes(data)
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                      
+                        slice_features = model.deit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+
+
+batch_size = 1
+
+# print (features_list)
+print (features_list[0].shape)
+
+# Create Dataset and DataLoader
+dataset = ADNIDataset(features_list, sex_encoded, age_list)
+train_size = int(0.8 * len(dataset))
+val_size = len(dataset) - train_size
+generator.manual_seed(universal_seed)
+train_dataset, val_dataset = torch.utils.data.random_split(dataset, [train_size, val_size], generator=generator)
+# Store the indices of the validation dataset
+val_indices = val_dataset.indices
+train_indices = train_dataset.indices
+
+train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=False)
+val_loader = DataLoader(val_dataset, batch_size=batch_size, shuffle=False)
+
+# Tracking outputs for validation samples
+val_outputs = {}
+train_outputs = {}
+
+# Initialize model, loss, and optimizer
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+
+###########################################
+# THIS IS WHERE YOU CHOOSE A MODEL TO TEST 
+###########################################
+
+import importlib
+
+# Assuming sys.argv[1] is the module name
+module_name = sys.argv[1]  # Example: "my_model"
+class_name = "AgePredictionCNN"  # The class you want to import
+
+try:
+    # Dynamically import the module
+    module = importlib.import_module(module_name)
+    
+    # Dynamically get the class
+    AgePredictionCNN = getattr(module, class_name)
+    
+    print(f"Successfully imported {class_name} from {module_name}.")
+
+except ImportError:
+    print(f"Module {module_name} could not be imported.")
+except AttributeError:
+    print(f"{class_name} does not exist in {module_name}.")
+
+##############################
+# MODEL IMPORTED DYNAMICALLY
+##############################
+
+print (features_list[0].shape)
+model = AgePredictionCNN((1, features_list[0].shape[0], features_list[0].shape[1])).to(device)
+criterion = nn.L1Loss()
+
+eval_crit = nn.L1Loss()
+# optimizer = optim.Adam(model.parameters(), lr=0.001)
+# adamw 
+optimizer = optim.AdamW(model.parameters(), lr=0.001, weight_decay=0.001)
+#optimizer = optim.Adam(model.parameters(), lr=0.0005, weight_decay=0.0001)
+best_loss = np.inf  # Initialize the best loss to infinity
+start_epoch = 0
+scheduler = optim.lr_scheduler.StepLR(optimizer, step_size=10, gamma=0.5)
+#scheduler = optim.lr_scheduler.MultiStepLR(optimizer, milestones=[10, 20, 30], gamma=0.5)
+#scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer, mode='min', factor=0.5, patience=5, verbose=True)
+#scheduler = optim.lr_scheduler.CosineAnnealingLR(optimizer, T_max=10, eta_min=0.0001, last_epoch=-1)
+#scheduler = optim.lr_scheduler.OneCycleLR(optimizer, max_lr=0.001, steps_per_epoch=len(train_loader), epochs=epochs)
+
+load_saved = sys.argv[2] # "last, "best"
+if load_saved != "none":
+    with open(f"model_dumps/mix/{model_name}/{slice_count}/{sys.argv[1]}_best_model_with_metadata.pkl", "rb") as f:
+        checkpoint = pickle.load(f)
+    best_loss = checkpoint["loss"]
+
+    # Load the checkpoint
+    with open(f"model_dumps/mix/{model_name}/{slice_count}/{sys.argv[1]}_{load_saved}_model_with_metadata.pkl", "rb") as f:
+        checkpoint = pickle.load(f)
+
+    # Restore model and optimizer state
+    model.load_state_dict(checkpoint["model_state"])
+    optimizer.load_state_dict(checkpoint["optimizer_state"])
+
+    # Restore RNG states
+    torch.set_rng_state(checkpoint["t_rng_st"])
+    np.random.set_state(checkpoint["n_rng_st"])
+    if torch.cuda.is_available() and checkpoint["cuda_rng_st"] is not None:
+        torch.cuda.set_rng_state_all(checkpoint["cuda_rng_st"])
+
+    # Retrieve metadata
+    start_epoch = checkpoint["epoch"] + 1
+    loaded_loss = checkpoint["loss"]
+
+    print(f"Loaded model from epoch {start_epoch} with validation loss {loaded_loss:.4f}, best loss {best_loss:.4f}")
+
+
+predicted_ages = None
+# Training loop
+epochs = int(sys.argv[3])
+
+# Initialize lists to track loss
+filename = sys.argv[1] 
+csv_file = f"model_dumps/mix/{model_name}/{slice_count}/{filename}.csv"
+
+# Load existing epoch data if the file exists
+if os.path.exists(csv_file):
+    epoch_data = pd.read_csv(csv_file).to_dict(orient="records")
+    print(f"Loaded existing epoch data from {csv_file}.")
+else:
+    epoch_data = []
+    print("No existing epoch data found. Starting fresh.")
+
+
+# Plot loss vs. epoch and save the figure
+def update_loss_plot(epoch_data, filename):
+    df = pd.DataFrame(epoch_data)
+    df.to_csv(f"model_dumps/mix/{slice_count}/{filename}.csv", index=False)  # Save the data to CSV
+    
+    plt.figure(figsize=(8, 6))
+    plt.plot(df['epoch'], df['train_loss'], label="Train Loss", marker="o")
+    plt.plot(df['epoch'], df['val_loss'], label="Validation Loss", marker="o")
+    plt.xlabel("Epoch")
+    plt.ylabel("Loss")
+    plt.title("Loss vs. Epoch")
+    plt.legend()
+    plt.grid(True)
+    plt.savefig(f"model_dumps/mix/{slice_count}/{filename}.png")
+    plt.close()
+
+# Training loop
+for epoch in range(start_epoch, epochs):
+    model.train()
+    train_loss = 0.0
+    predicted_ages = []
+
+    for idx, (features, sex, age) in enumerate(train_loader):
+        features = features.unsqueeze(1).to(device)  # Add channel dimension
+        sex = sex.to(device)
+        age = age.to(device)
+        optimizer.zero_grad()
+        outputs = model(features, sex)
+        # Store the output for each sample in the batch
+        for i in range(outputs.size(0)):
+            train_outputs[train_indices[idx * batch_size + i]] = outputs[i].item()
+
+        loss = criterion(outputs.squeeze(), age)
+        loss.backward()
+        optimizer.step()
+
+        train_loss += loss.item()
+
+    train_loss /= len(train_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Train Loss: {train_loss:.4f}")
+
+    # Validation loop
+    model.eval()
+    val_loss = 0.0
+    with torch.no_grad():
+        for idx, (features, sex, age) in enumerate(val_loader):
+            features = features.unsqueeze(1).to(device)
+            sex = sex.to(device)
+            age = age.to(device)
+
+            outputs = model(features, sex)
+            loss = eval_crit(outputs.squeeze(), age)
+            val_loss += loss.item()
+
+            # Save the predicted age for the current validation sample
+            for i in range(outputs.size(0)):
+                val_outputs[val_indices[idx * batch_size + i]] = outputs[i].item()
+            # val_outputs[val_indices[idx]] = outputs.item()
+
+    val_loss /= len(val_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Validation Loss: {val_loss:.4f}")
+
+    #scheduler.step()
+
+    # Save the last model with metadata
+    print(f"Saving last model...")
+    checkpoint = {
+        "model_state": model.state_dict(),
+        "optimizer_state": optimizer.state_dict(),
+        "epoch": epoch,
+        "loss": val_loss,
+        "t_rng_st": torch.get_rng_state(),
+        "n_rng_st": np.random.get_state(),
+        "cuda_rng_st": torch.cuda.get_rng_state_all() if torch.cuda.is_available() else None
+    }
+    with open(f"model_dumps/mix/{model_name}/{slice_count}/{sys.argv[1]}_last_model_with_metadata.pkl", "wb") as f:
+        pickle.dump(checkpoint, f)
+    print(f"Last model saved...")
+
+    # Check if validation loss improved
+    if val_loss < best_loss:
+        best_loss = val_loss
+        print(f"Validation loss improved to {best_loss:.4f}. Saving model...")
+        with open(f"model_dumps/mix/{model_name}/{slice_count}/{sys.argv[1]}_best_model_with_metadata.pkl", "wb") as f:
+            pickle.dump(checkpoint, f)
+
+    # Save predictions and create DataFrames (same as before)
+    # ...
+
+    # Update epoch data and save the loss plot
+    epoch_data.append({
+        "epoch": epoch + 1,
+        "train_loss": train_loss,
+        "val_loss": val_loss
+    })
+    update_loss_plot(epoch_data, sys.argv[1])
+
+    max_index = max(train_outputs.keys())
+    # Create a DataFrame with NaN for all indices initially
+    df_trn = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
+    # Assign the values to their respective indices
+    for index, value in train_outputs.items():
+        df_trn.loc[index, "Predicted_Age"] = value
+    # print (df_trn)
+
+    df2 = df.copy()
+    df2['Predicted_Age'] = df_trn['Predicted_Age']
+    train_df = df2.loc[train_outputs.keys()]
+    # print (train_df)
+    train_df.to_csv(f"model_dumps/mix/{model_name}/{slice_count}/{sys.argv[1]}_predicted_ages_train.csv")
+
+    max_index = max(val_outputs.keys())
+    # Create a DataFrame with NaN for all indices initially
+    df_pred = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
+    # Assign the values to their respective indices
+    for index, value in val_outputs.items():
+        df_pred.loc[index, "Predicted_Age"] = value
+    # print (df_pred)
+
+    df1 = df.copy()
+    df1['Predicted_Age'] = df_pred['Predicted_Age']
+    test_df = df1.loc[val_outputs.keys()]
+    print (test_df)
+    test_df.to_csv(f"model_dumps/mix/{model_name}/{slice_count}/{sys.argv[1]}_predicted_ages_val.csv")
+
+
+    # Map unique first 4 characters of ImageID to color codes
+    unique_groups = test_df['ImageID'].str[:3].unique()
+    group_to_color = {group: i for i, group in enumerate(unique_groups)}
+
+    # Assign colors based on the mapping
+    cmap = plt.get_cmap('tab10')  # Change colormap as desired
+    colors = [cmap(group_to_color[group]) for group in test_df['ImageID'].str[:3]]
+
+    # Check that the predictions have been added to the DataFrame
+    # Plot Age vs. Predicted Age
+    plt.figure(figsize=(8, 6))
+    plt.scatter(test_df['Age'], test_df['Predicted_Age'], color=colors, label='Predicted vs Actual')
+    # plt.plot(test_df['Age'], test_df['Age'], color='red', linestyle='--', label='Perfect Prediction')  # Optional: Line of perfect prediction
+    # Add legend for colors based on ImageID groups
+    handles = [plt.Line2D([0], [0], marker='o', color=cmap(i), linestyle='', markersize=10) 
+            for i, group in enumerate(unique_groups)]
+    plt.legend(handles, unique_groups, title="ImageID Groups")
+    plt.xlabel('Age')
+    plt.ylabel('Predicted Age')
+    plt.title('Age vs Predicted Age')
+    plt.grid(True)
+    plt.savefig(f"model_dumps/mix/{model_name}/{slice_count}/plots/{model_name}_cnn_{sys.argv[1]}_epoch{epoch}.png")
diff --git a/deit_train_metrics_32.csv b/deit_train_metrics_32.csv
new file mode 100644
index 0000000..fa2d6f5
--- /dev/null
+++ b/deit_train_metrics_32.csv
@@ -0,0 +1,2 @@
+1,1.8282773258434377,33.88637367491166
+2,1.3353317810762213,50.72603798586572
diff --git a/oasis.sh b/oasis.sh
new file mode 100644
index 0000000..7f0aa93
--- /dev/null
+++ b/oasis.sh
@@ -0,0 +1,37 @@
+#!/bin/bash
+
+# Define the base directory and output directory
+BASE_DIR="./OASIS"
+OUTPUT_DIR="./OASIS/oasis_merged"
+
+# Create the output directory if it does not exist
+mkdir -p "$OUTPUT_DIR"
+
+# Loop through disc1 to disc12
+for disc in {1..12}; do
+    # Find all .hdr files in the specific folder structure
+    find "$BASE_DIR" -type f -path "*/disc${disc}/*/PROCESSED/MPRAGE/SUBJ_111/*.hdr" | while read -r hdr_file; do
+        # Get the corresponding .img file
+        img_file="${hdr_file%.hdr}.img"
+
+        # Check if the corresponding .img file exists
+        if [ -f "$img_file" ]; then
+            # Extract the base name without extension
+            base_name=$(basename "$hdr_file" .hdr)
+
+            # Define the output file path
+            output_file="$OUTPUT_DIR/${base_name}.nii.gz"
+
+            # Run the conversion using fslchfiletype
+            fslchfiletype NIFTI_GZ "$hdr_file" "$output_file"
+
+            # Log the operation
+            echo "Converted: $hdr_file to $output_file"
+        else
+            echo "Missing .img file for: $hdr_file" >&2
+        fi
+    done
+done
+
+# Notify completion
+echo "Conversion completed. All .nii.gz files are saved in $OUTPUT_DIR."
diff --git a/roi.py b/roi.py
new file mode 100644
index 0000000..3061fd5
--- /dev/null
+++ b/roi.py
@@ -0,0 +1,62 @@
+import os
+import glob
+import nibabel as nib
+import numpy as np
+
+def main():
+    # Paths to dataset folders
+    dataset_path = "ixi"
+    output_folder = "ixiclroi"    
+    # Create the output folder if it doesn't exist
+    os.makedirs(output_folder, exist_ok=True)
+
+    # Retrieve file paths for brain volumes (supporting .nii.gz)
+    brain_volume_list = sorted(glob.glob(os.path.join(dataset_path, '**/*.nii.gz'), recursive=True))
+
+    # Crop and save volumes
+    crop_and_save_brain_volumes(brain_volume_list, output_folder)
+
+    print("Cropped brain volumes saved to:", output_folder)
+
+
+def calculate_bounding_box_from_volume(volume):
+    # Find indices of non-zero values
+    non_zero_indices = np.argwhere(volume > 0)
+
+    # Calculate min and max indices along each dimension
+    min_indices = np.min(non_zero_indices, axis=0)
+    max_indices = np.max(non_zero_indices, axis=0)
+
+    # Convert indices to integers
+    min_indices = min_indices.astype(int)
+    max_indices = max_indices.astype(int)
+
+    return min_indices, max_indices
+
+
+def crop_and_save_brain_volumes(brain_volume_list, output_folder):
+    for brain_path in brain_volume_list:
+        # Load the brain volume
+        brain_volume = nib.load(brain_path)
+        brain_data = brain_volume.get_fdata()
+
+        # Calculate bounding box from the brain volume
+        min_indices, max_indices = calculate_bounding_box_from_volume(brain_data)
+
+        # Crop the volume
+        cropped_brain = brain_data[min_indices[0]:max_indices[0] + 1,
+                                   min_indices[1]:max_indices[1] + 1,
+                                   min_indices[2]:max_indices[2] + 1]
+
+        # Save the cropped brain volume
+        _, filename = os.path.split(brain_path)
+        filename_without_extension = os.path.splitext(os.path.splitext(filename)[0])[0]  # Handles .nii.gz
+        cropped_output_path = os.path.join(output_folder, filename_without_extension + "_cropped.nii.gz")
+
+        nib.save(nib.Nifti1Image(cropped_brain, brain_volume.affine), cropped_output_path)
+
+        print(f"Processed and saved: {filename}")
+
+
+if __name__ == "__main__":
+    main()
diff --git a/train_dataset.csv b/train_dataset.csv
new file mode 100644
index 0000000..0177cbe
--- /dev/null
+++ b/train_dataset.csv
@@ -0,0 +1,1785 @@
+ImageID,Sex,Age,filepath
+ADNI1283849,F,62.4,adni_storage/ADNI_nii_gz_bias_corrected/I1283849_305_S_6845.stripped.N4.nii.gz
+ADNI308453,M,61.3,adni_storage/ADNI_nii_gz_bias_corrected/I308453_073_S_4795.stripped.N4.nii.gz
+fcon1000Milwaukee_sub61779,F,56.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub61779.stripped.N4.nii.gz
+fcon1000Beijing_sub67844,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub67844.stripped.N4.nii.gz
+ADNI997517,F,57.9,adni_storage/ADNI_nii_gz_bias_corrected/I997517_131_S_6170.stripped.N4.nii.gz
+DLBS0028440_session_1,F,34.25,dlbs_storage/DLBS_bias_corrected/0028440_session_1.stripped.N4.nii.gz
+sub-CC520279,F,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520279.stripped.N4.nii.gz
+sub-CC610653,M,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610653.stripped.N4.nii.gz
+fcon1000AnnArbor_sub18698,M,29.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub18698.stripped.N4.nii.gz
+DLBS0028358_session_1,M,66.63,dlbs_storage/DLBS_bias_corrected/0028358_session_1.stripped.N4.nii.gz
+sub-CC420587,F,49.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420587.stripped.N4.nii.gz
+sub-CC520147,F,61.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520147.stripped.N4.nii.gz
+fcon1000Cambridge_sub13216,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub13216.stripped.N4.nii.gz
+sub-CC410091,F,57.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410091.stripped.N4.nii.gz
+ADNI341152,M,60.6,adni_storage/ADNI_nii_gz_bias_corrected/I341152_128_S_4599.stripped.N4.nii.gz
+fcon1000Newark_sub41006,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub41006.stripped.N4.nii.gz
+DLBS0028376_session_1,F,70.6,dlbs_storage/DLBS_bias_corrected/0028376_session_1.stripped.N4.nii.gz
+ADNI363016,F,61.4,adni_storage/ADNI_nii_gz_bias_corrected/I363016_036_S_4894.stripped.N4.nii.gz
+sub-CC510076,M,62.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510076.stripped.N4.nii.gz
+fcon1000Cambridge_sub04270,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub04270.stripped.N4.nii.gz
+ADNI1469766,F,58.6,adni_storage/ADNI_nii_gz_bias_corrected/I1469766_073_S_6670.stripped.N4.nii.gz
+fcon1000Berlin_sub91966,F,27.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub91966.stripped.N4.nii.gz
+fcon1000Cambridge_sub15432,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub15432.stripped.N4.nii.gz
+fcon1000AnnArbor_sub96621,M,15.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub96621.stripped.N4.nii.gz
+fcon1000Beijing_sub58614,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub58614.stripped.N4.nii.gz
+sub-CC420149,M,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420149.stripped.N4.nii.gz
+fcon1000AnnArbor_sub90127,M,77.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub90127.stripped.N4.nii.gz
+DLBS0028633_session_1,M,46.44,dlbs_storage/DLBS_bias_corrected/0028633_session_1.stripped.N4.nii.gz
+DLBS0028565_session_1,M,80.84,dlbs_storage/DLBS_bias_corrected/0028565_session_1.stripped.N4.nii.gz
+sub-CC320202,M,45.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320202.stripped.N4.nii.gz
+ADNI999074,M,62.0,adni_storage/ADNI_nii_gz_bias_corrected/I999074_130_S_6329.stripped.N4.nii.gz
+ADNI1490106,M,57.8,adni_storage/ADNI_nii_gz_bias_corrected/I1490106_037_S_6993.stripped.N4.nii.gz
+ADNI10453857,F,58.6,adni_storage/ADNI_nii_gz_bias_corrected/I10453857_037_S_10086.stripped.N4.nii.gz
+ADNI313319,F,56.5,adni_storage/ADNI_nii_gz_bias_corrected/I313319_128_S_4607.stripped.N4.nii.gz
+fcon1000Milwaukee_sub76378,F,52.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub76378.stripped.N4.nii.gz
+fcon1000AnnArbor_sub16960,M,13.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub16960.stripped.N4.nii.gz
+sub-CC520078,M,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520078.stripped.N4.nii.gz
+fcon1000Cleveland_sub80263,F,44.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub80263.stripped.N4.nii.gz
+DLBS0028497_session_1,F,49.13,dlbs_storage/DLBS_bias_corrected/0028497_session_1.stripped.N4.nii.gz
+sub-CC620793,M,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620793.stripped.N4.nii.gz
+fcon1000Cleveland_sub12330,F,54.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub12330.stripped.N4.nii.gz
+ADNI341366,M,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I341366_116_S_4625.stripped.N4.nii.gz
+ADNI90756,F,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I90756_051_S_1072.stripped.N4.nii.gz
+ADNI308459,M,61.3,adni_storage/ADNI_nii_gz_bias_corrected/I308459_073_S_4795.stripped.N4.nii.gz
+sub-CC520477,M,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520477.stripped.N4.nii.gz
+DLBS0028620_session_1,F,70.66,dlbs_storage/DLBS_bias_corrected/0028620_session_1.stripped.N4.nii.gz
+ADNI98971,M,63.9,adni_storage/ADNI_nii_gz_bias_corrected/I98971_016_S_0769.stripped.N4.nii.gz
+ADNI1639219,M,57.4,adni_storage/ADNI_nii_gz_bias_corrected/I1639219_011_S_7112.stripped.N4.nii.gz
+sub-CC110069,F,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110069.stripped.N4.nii.gz
+fcon1000Beijing_sub39725,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub39725.stripped.N4.nii.gz
+DLBS0028585_session_1,F,31.4,dlbs_storage/DLBS_bias_corrected/0028585_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub87745,M,14.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub87745.stripped.N4.nii.gz
+fcon1000Newark_sub86204,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub86204.stripped.N4.nii.gz
+sub-CC510323,F,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510323.stripped.N4.nii.gz
+fcon1000Cambridge_sub62383,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub62383.stripped.N4.nii.gz
+sub-CC120234,F,24.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120234.stripped.N4.nii.gz
+ADNI1495816,M,59.3,adni_storage/ADNI_nii_gz_bias_corrected/I1495816_003_S_7010.stripped.N4.nii.gz
+sub-CC310008,M,45.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310008.stripped.N4.nii.gz
+sub-CC320553,F,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320553.stripped.N4.nii.gz
+DLBS0028450_session_1,F,44.99,dlbs_storage/DLBS_bias_corrected/0028450_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub18758,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub18758.stripped.N4.nii.gz
+ADNI984775,F,59.1,adni_storage/ADNI_nii_gz_bias_corrected/I984775_305_S_6313.stripped.N4.nii.gz
+sub-CC721377,M,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721377.stripped.N4.nii.gz
+sub-CC510050,F,59.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510050.stripped.N4.nii.gz
+ADNI1568731,F,56.3,adni_storage/ADNI_nii_gz_bias_corrected/I1568731_067_S_7063.stripped.N4.nii.gz
+fcon1000Cambridge_sub54846,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub54846.stripped.N4.nii.gz
+sub-CC223286,M,36.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC223286.stripped.N4.nii.gz
+sub-CC121106,F,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121106.stripped.N4.nii.gz
+ADNI10888787,F,59.2,adni_storage/ADNI_nii_gz_bias_corrected/I10888787_005_S_10101.stripped.N4.nii.gz
+ADNI1315300,F,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I1315300_305_S_6313.stripped.N4.nii.gz
+fcon1000Beijing_sub20765,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub20765.stripped.N4.nii.gz
+fcon1000Beijing_sub73245,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub73245.stripped.N4.nii.gz
+fcon1000Beijing_sub80927,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub80927.stripped.N4.nii.gz
+sub-CC220372,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220372.stripped.N4.nii.gz
+fcon1000Cleveland_sub99664,F,57.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub99664.stripped.N4.nii.gz
+fcon1000Milwaukee_sub91468,M,44.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub91468.stripped.N4.nii.gz
+fcon1000Cambridge_sub83683,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub83683.stripped.N4.nii.gz
+sub-CC520254,F,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520254.stripped.N4.nii.gz
+DLBS0028422_session_1,F,32.91,dlbs_storage/DLBS_bias_corrected/0028422_session_1.stripped.N4.nii.gz
+sub-CC320160,F,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320160.stripped.N4.nii.gz
+DLBS0028612_session_1,F,67.61,dlbs_storage/DLBS_bias_corrected/0028612_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub74386,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub74386.stripped.N4.nii.gz
+ADNI10927885,F,57.1,adni_storage/ADNI_nii_gz_bias_corrected/I10927885_035_S_10193.stripped.N4.nii.gz
+sub-CC120120,M,25.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120120.stripped.N4.nii.gz
+fcon1000NewYork_sub20676,M,49.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub20676.stripped.N4.nii.gz
+fcon1000Milwaukee_sub67948,M,48.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub67948.stripped.N4.nii.gz
+ADNI1038250,F,60.3,adni_storage/ADNI_nii_gz_bias_corrected/I1038250_067_S_6529.stripped.N4.nii.gz
+sub-CC420582,M,54.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420582.stripped.N4.nii.gz
+fcon1000Cambridge_sub47498,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub47498.stripped.N4.nii.gz
+DLBS0028427_session_1,M,25.7,dlbs_storage/DLBS_bias_corrected/0028427_session_1.stripped.N4.nii.gz
+DLBS0028328_session_1,M,30.47,dlbs_storage/DLBS_bias_corrected/0028328_session_1.stripped.N4.nii.gz
+ADNI10883231,F,56.4,adni_storage/ADNI_nii_gz_bias_corrected/I10883231_941_S_10169.stripped.N4.nii.gz
+fcon1000Bangor_sub66585,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub66585.stripped.N4.nii.gz
+fcon1000NewYork_sub22608,F,23.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub22608.stripped.N4.nii.gz
+fcon1000Berlin_sub12855,M,33.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub12855.stripped.N4.nii.gz
+DLBS0028436_session_1,F,75.99,dlbs_storage/DLBS_bias_corrected/0028436_session_1.stripped.N4.nii.gz
+sub-CC420217,M,50.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420217.stripped.N4.nii.gz
+ADNI1221673,F,63.4,adni_storage/ADNI_nii_gz_bias_corrected/I1221673_013_S_6780.stripped.N4.nii.gz
+sub-CC721504,M,82.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721504.stripped.N4.nii.gz
+sub-CC620490,F,74.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620490.stripped.N4.nii.gz
+fcon1000Milwaukee_sub44912,F,49.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub44912.stripped.N4.nii.gz
+fcon1000Milwaukee_sub56084,M,53.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub56084.stripped.N4.nii.gz
+fcon1000Newark_sub49705,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub49705.stripped.N4.nii.gz
+sub-CC410179,M,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410179.stripped.N4.nii.gz
+DLBS0028460_session_1,F,37.05,dlbs_storage/DLBS_bias_corrected/0028460_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub92288,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub92288.stripped.N4.nii.gz
+ADNI308627,M,57.0,adni_storage/ADNI_nii_gz_bias_corrected/I308627_014_S_4039.stripped.N4.nii.gz
+fcon1000NewYork_sub16607,M,21.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub16607.stripped.N4.nii.gz
+sub-CC711027,M,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC711027.stripped.N4.nii.gz
+ADNI1003993,F,64.1,adni_storage/ADNI_nii_gz_bias_corrected/I1003993_023_S_6399.stripped.N4.nii.gz
+sub-CC420226,M,50.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420226.stripped.N4.nii.gz
+DLBS0028458_session_1,F,35.43,dlbs_storage/DLBS_bias_corrected/0028458_session_1.stripped.N4.nii.gz
+fcon1000Bangor_sub91556,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub91556.stripped.N4.nii.gz
+ADNI99327,M,61.1,adni_storage/ADNI_nii_gz_bias_corrected/I99327_020_S_1288.stripped.N4.nii.gz
+fcon1000Cambridge_sub58360,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub58360.stripped.N4.nii.gz
+ADNI238745,M,56.0,adni_storage/ADNI_nii_gz_bias_corrected/I238745_014_S_4039.stripped.N4.nii.gz
+fcon1000Cambridge_sub27613,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub27613.stripped.N4.nii.gz
+DLBS0028547_session_1,F,32.13,dlbs_storage/DLBS_bias_corrected/0028547_session_1.stripped.N4.nii.gz
+fcon1000NewYork_sub17109,M,46.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub17109.stripped.N4.nii.gz
+sub-CC420004,F,49.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420004.stripped.N4.nii.gz
+sub-CC720071,M,82.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720071.stripped.N4.nii.gz
+sub-CC721585,F,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721585.stripped.N4.nii.gz
+fcon1000Cambridge_sub10619,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub10619.stripped.N4.nii.gz
+DLBS0028555_session_1,M,64.86,dlbs_storage/DLBS_bias_corrected/0028555_session_1.stripped.N4.nii.gz
+ADNI10968674,F,59.2,adni_storage/ADNI_nii_gz_bias_corrected/I10968674_052_S_10252.stripped.N4.nii.gz
+sub-CC620406,M,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620406.stripped.N4.nii.gz
+fcon1000Beijing_sub10186,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub10186.stripped.N4.nii.gz
+fcon1000Atlanta_sub49816,M,27.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub49816.stripped.N4.nii.gz
+sub-CC520424,M,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520424.stripped.N4.nii.gz
+fcon1000Berlin_sub38279,M,29.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub38279.stripped.N4.nii.gz
+ADNI10277636,F,63.6,adni_storage/ADNI_nii_gz_bias_corrected/I10277636_033_S_10005.stripped.N4.nii.gz
+fcon1000Beijing_sub34943,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub34943.stripped.N4.nii.gz
+fcon1000Cambridge_sub90674,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub90674.stripped.N4.nii.gz
+fcon1000Beijing_sub85818,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub85818.stripped.N4.nii.gz
+DLBS0028357_session_1,F,56.36,dlbs_storage/DLBS_bias_corrected/0028357_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub23869,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub23869.stripped.N4.nii.gz
+ADNI280812,F,63.9,adni_storage/ADNI_nii_gz_bias_corrected/I280812_023_S_4448.stripped.N4.nii.gz
+sub-CC520097,F,62.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520097.stripped.N4.nii.gz
+sub-CC620085,F,70.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620085.stripped.N4.nii.gz
+DLBS0028570_session_1,F,63.31,dlbs_storage/DLBS_bias_corrected/0028570_session_1.stripped.N4.nii.gz
+sub-CC610146,F,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610146.stripped.N4.nii.gz
+sub-CC620919,M,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620919.stripped.N4.nii.gz
+sub-CC720941,M,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720941.stripped.N4.nii.gz
+ADNI423071,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I423071_128_S_4607.stripped.N4.nii.gz
+fcon1000Beijing_sub76377,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub76377.stripped.N4.nii.gz
+DLBS0028444_session_1,F,81.41,dlbs_storage/DLBS_bias_corrected/0028444_session_1.stripped.N4.nii.gz
+ADNI290867,M,60.0,adni_storage/ADNI_nii_gz_bias_corrected/I290867_128_S_4599.stripped.N4.nii.gz
+fcon1000Baltimore_sub54329,F,30.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub54329.stripped.N4.nii.gz
+ADNI998806,F,62.0,adni_storage/ADNI_nii_gz_bias_corrected/I998806_037_S_6271.stripped.N4.nii.gz
+ADNI285070,M,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I285070_073_S_4552.stripped.N4.nii.gz
+sub-CC520083,F,65.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520083.stripped.N4.nii.gz
+DLBS0028552_session_1,F,48.86,dlbs_storage/DLBS_bias_corrected/0028552_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub42253,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub42253.stripped.N4.nii.gz
+fcon1000Beijing_sub75878,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub75878.stripped.N4.nii.gz
+ADNI1611592,M,58.2,adni_storage/ADNI_nii_gz_bias_corrected/I1611592_033_S_6976.stripped.N4.nii.gz
+ADNI352948,M,64.0,adni_storage/ADNI_nii_gz_bias_corrected/I352948_002_S_1155.stripped.N4.nii.gz
+fcon1000Cambridge_sub35430,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub35430.stripped.N4.nii.gz
+fcon1000Cambridge_sub23780,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub23780.stripped.N4.nii.gz
+sub-CC110182,F,18.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110182.stripped.N4.nii.gz
+fcon1000Atlanta_sub72971,M,30.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub72971.stripped.N4.nii.gz
+ADNI10283161,M,59.2,adni_storage/ADNI_nii_gz_bias_corrected/I10283161_941_S_6998.stripped.N4.nii.gz
+sub-CC420089,F,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420089.stripped.N4.nii.gz
+sub-CC711244,F,85.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC711244.stripped.N4.nii.gz
+fcon1000Cleveland_sub61868,M,48.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub61868.stripped.N4.nii.gz
+ADNI1184048,F,62.6,adni_storage/ADNI_nii_gz_bias_corrected/I1184048_022_S_6013.stripped.N4.nii.gz
+ADNI966268,F,62.1,adni_storage/ADNI_nii_gz_bias_corrected/I966268_020_S_6227.stripped.N4.nii.gz
+fcon1000Cambridge_sub62908,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub62908.stripped.N4.nii.gz
+fcon1000Beijing_sub96163,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub96163.stripped.N4.nii.gz
+fcon1000Munchen_sub36052,F,63.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub36052.stripped.N4.nii.gz
+ADNI1067189,M,64.1,adni_storage/ADNI_nii_gz_bias_corrected/I1067189_012_S_6073.stripped.N4.nii.gz
+DLBS0028424_session_1,F,77.79,dlbs_storage/DLBS_bias_corrected/0028424_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub04665,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub04665.stripped.N4.nii.gz
+fcon1000Berlin_sub06716,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub06716.stripped.N4.nii.gz
+ADNI10883247,M,60.0,adni_storage/ADNI_nii_gz_bias_corrected/I10883247_037_S_10117.stripped.N4.nii.gz
+sub-CC322186,M,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC322186.stripped.N4.nii.gz
+sub-CC410243,F,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410243.stripped.N4.nii.gz
+DLBS0028336_session_1,F,80.65,dlbs_storage/DLBS_bias_corrected/0028336_session_1.stripped.N4.nii.gz
+ADNI263858,F,56.5,adni_storage/ADNI_nii_gz_bias_corrected/I263858_031_S_4024.stripped.N4.nii.gz
+sub-CC320616,M,39.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320616.stripped.N4.nii.gz
+fcon1000Cambridge_sub27230,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub27230.stripped.N4.nii.gz
+sub-CC120166,M,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120166.stripped.N4.nii.gz
+sub-CC610658,M,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610658.stripped.N4.nii.gz
+ADNI1608103,M,63.4,adni_storage/ADNI_nii_gz_bias_corrected/I1608103_016_S_6941.stripped.N4.nii.gz
+ADNI70745,M,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I70745_016_S_0769.stripped.N4.nii.gz
+ADNI10546533,F,58.6,adni_storage/ADNI_nii_gz_bias_corrected/I10546533_137_S_6906.stripped.N4.nii.gz
+fcon1000Beijing_sub07716,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub07716.stripped.N4.nii.gz
+fcon1000Beijing_sub33943,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub33943.stripped.N4.nii.gz
+sub-CC720329,M,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720329.stripped.N4.nii.gz
+ADNI1488791,F,54.9,adni_storage/ADNI_nii_gz_bias_corrected/I1488791_037_S_6951.stripped.N4.nii.gz
+ADNI270530,M,56.5,adni_storage/ADNI_nii_gz_bias_corrected/I270530_014_S_4039.stripped.N4.nii.gz
+fcon1000Cambridge_sub52442,F,25.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub52442.stripped.N4.nii.gz
+sub-CC510392,M,67.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510392.stripped.N4.nii.gz
+sub-CC222120,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222120.stripped.N4.nii.gz
+fcon1000Atlanta_sub18219,F,33.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub18219.stripped.N4.nii.gz
+ADNI417668,M,62.0,adni_storage/ADNI_nii_gz_bias_corrected/I417668_128_S_4599.stripped.N4.nii.gz
+sub-CC520745,F,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520745.stripped.N4.nii.gz
+ADNI879209,M,56.0,adni_storage/ADNI_nii_gz_bias_corrected/I879209_114_S_6039.stripped.N4.nii.gz
+fcon1000Beijing_sub36580,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub36580.stripped.N4.nii.gz
+fcon1000Cambridge_sub50953,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub50953.stripped.N4.nii.gz
+fcon1000AnnArbor_sub62819,M,65.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub62819.stripped.N4.nii.gz
+sub-CC222797,F,31.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222797.stripped.N4.nii.gz
+DLBS0028546_session_1,M,21.71,dlbs_storage/DLBS_bias_corrected/0028546_session_1.stripped.N4.nii.gz
+sub-CC112141,M,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC112141.stripped.N4.nii.gz
+fcon1000Beijing_sub73279,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub73279.stripped.N4.nii.gz
+sub-CC621080,M,70.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC621080.stripped.N4.nii.gz
+fcon1000Cambridge_sub77435,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub77435.stripped.N4.nii.gz
+sub-CC310402,F,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310402.stripped.N4.nii.gz
+fcon1000Beijing_sub28206,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub28206.stripped.N4.nii.gz
+fcon1000Munchen_sub28697,F,73.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub28697.stripped.N4.nii.gz
+DLBS0028487_session_1,F,73.56,dlbs_storage/DLBS_bias_corrected/0028487_session_1.stripped.N4.nii.gz
+sub-CC710566,M,83.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710566.stripped.N4.nii.gz
+fcon1000Cambridge_sub93609,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub93609.stripped.N4.nii.gz
+sub-CC121158,F,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121158.stripped.N4.nii.gz
+sub-CC620279,M,77.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620279.stripped.N4.nii.gz
+fcon1000Baltimore_sub23750,M,37.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub23750.stripped.N4.nii.gz
+sub-CC310142,M,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310142.stripped.N4.nii.gz
+ADNI442428,F,59.8,adni_storage/ADNI_nii_gz_bias_corrected/I442428_041_S_4874.stripped.N4.nii.gz
+DLBS0028337_session_1,F,53.34,dlbs_storage/DLBS_bias_corrected/0028337_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub11072,F,25.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub11072.stripped.N4.nii.gz
+ADNI10461519,F,61.9,adni_storage/ADNI_nii_gz_bias_corrected/I10461519_022_S_10043.stripped.N4.nii.gz
+fcon1000NewYork_sub20691,M,26.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub20691.stripped.N4.nii.gz
+fcon1000Beijing_sub48501,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub48501.stripped.N4.nii.gz
+sub-CC220223,M,33.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220223.stripped.N4.nii.gz
+ADNI907713,F,58.7,adni_storage/ADNI_nii_gz_bias_corrected/I907713_014_S_6076.stripped.N4.nii.gz
+fcon1000AnnArbor_sub51248,M,40.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub51248.stripped.N4.nii.gz
+DLBS0028483_session_1,F,47.0,dlbs_storage/DLBS_bias_corrected/0028483_session_1.stripped.N4.nii.gz
+fcon1000Newark_sub59397,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub59397.stripped.N4.nii.gz
+fcon1000Beijing_sub73098,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub73098.stripped.N4.nii.gz
+sub-CC420566,F,50.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420566.stripped.N4.nii.gz
+fcon1000Berlin_sub91116,M,39.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub91116.stripped.N4.nii.gz
+sub-CC620466,M,75.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620466.stripped.N4.nii.gz
+sub-CC310463,F,45.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310463.stripped.N4.nii.gz
+ADNI1438743,F,64.3,adni_storage/ADNI_nii_gz_bias_corrected/I1438743_114_S_6917.stripped.N4.nii.gz
+DLBS0028389_session_1,M,33.45,dlbs_storage/DLBS_bias_corrected/0028389_session_1.stripped.N4.nii.gz
+ADNI386092,F,58.8,adni_storage/ADNI_nii_gz_bias_corrected/I386092_041_S_4874.stripped.N4.nii.gz
+ADNI341369,M,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I341369_116_S_4625.stripped.N4.nii.gz
+sub-CC721107,F,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721107.stripped.N4.nii.gz
+fcon1000Cambridge_sub21755,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub21755.stripped.N4.nii.gz
+sub-CC520211,F,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520211.stripped.N4.nii.gz
+sub-CC220999,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220999.stripped.N4.nii.gz
+sub-CC710446,M,85.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710446.stripped.N4.nii.gz
+fcon1000Beijing_sub82714,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub82714.stripped.N4.nii.gz
+ADNI296876,F,58.5,adni_storage/ADNI_nii_gz_bias_corrected/I296876_012_S_4642.stripped.N4.nii.gz
+fcon1000Beijing_sub30556,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub30556.stripped.N4.nii.gz
+fcon1000Beijing_sub28801,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub28801.stripped.N4.nii.gz
+ADNI86563,F,57.3,adni_storage/ADNI_nii_gz_bias_corrected/I86563_016_S_1121.stripped.N4.nii.gz
+fcon1000Beijing_sub17688,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub17688.stripped.N4.nii.gz
+sub-CC610178,M,72.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610178.stripped.N4.nii.gz
+ADNI1577482,F,56.7,adni_storage/ADNI_nii_gz_bias_corrected/I1577482_305_S_6899.stripped.N4.nii.gz
+DLBS0028475_session_1,M,54.05,dlbs_storage/DLBS_bias_corrected/0028475_session_1.stripped.N4.nii.gz
+ADNI271073,M,62.6,adni_storage/ADNI_nii_gz_bias_corrected/I271073_116_S_4195.stripped.N4.nii.gz
+DLBS0028531_session_1,F,21.27,dlbs_storage/DLBS_bias_corrected/0028531_session_1.stripped.N4.nii.gz
+sub-CC120550,M,19.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120550.stripped.N4.nii.gz
+fcon1000Cambridge_sub73399,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub73399.stripped.N4.nii.gz
+fcon1000Beijing_sub86114,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub86114.stripped.N4.nii.gz
+fcon1000Beijing_sub62966,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub62966.stripped.N4.nii.gz
+DLBS0028384_session_1,F,30.5,dlbs_storage/DLBS_bias_corrected/0028384_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub66781,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub66781.stripped.N4.nii.gz
+sub-CC221733,M,36.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221733.stripped.N4.nii.gz
+fcon1000Beijing_sub66158,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub66158.stripped.N4.nii.gz
+sub-CC610344,F,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610344.stripped.N4.nii.gz
+fcon1000Milwaukee_sub49975,M,46.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub49975.stripped.N4.nii.gz
+DLBS0028332_session_1,M,71.29,dlbs_storage/DLBS_bias_corrected/0028332_session_1.stripped.N4.nii.gz
+DLBS0028471_session_1,F,67.47,dlbs_storage/DLBS_bias_corrected/0028471_session_1.stripped.N4.nii.gz
+ADNI345557,F,63.0,adni_storage/ADNI_nii_gz_bias_corrected/I345557_019_S_5019.stripped.N4.nii.gz
+fcon1000Leiden_sub52922,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub52922.stripped.N4.nii.gz
+sub-CC210250,M,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210250.stripped.N4.nii.gz
+ADNI1092245,M,57.4,adni_storage/ADNI_nii_gz_bias_corrected/I1092245_114_S_6039.stripped.N4.nii.gz
+fcon1000Newark_sub43517,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub43517.stripped.N4.nii.gz
+fcon1000Bangor_sub00031,M,29.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub00031.stripped.N4.nii.gz
+fcon1000NewYork_sub46856,F,8.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub46856.stripped.N4.nii.gz
+ADNI1032324,F,56.5,adni_storage/ADNI_nii_gz_bias_corrected/I1032324_070_S_6548.stripped.N4.nii.gz
+sub-CC610594,F,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610594.stripped.N4.nii.gz
+fcon1000Munchen_sub09035,M,70.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub09035.stripped.N4.nii.gz
+fcon1000Beijing_sub65659,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub65659.stripped.N4.nii.gz
+fcon1000Cambridge_sub05453,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub05453.stripped.N4.nii.gz
+fcon1000Cambridge_sub51050,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub51050.stripped.N4.nii.gz
+sub-CC410173,M,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410173.stripped.N4.nii.gz
+DLBS0028407_session_1,M,86.86,dlbs_storage/DLBS_bias_corrected/0028407_session_1.stripped.N4.nii.gz
+ADNI1304804,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I1304804_168_S_6085.stripped.N4.nii.gz
+ADNI365456,M,62.1,adni_storage/ADNI_nii_gz_bias_corrected/I365456_073_S_4795.stripped.N4.nii.gz
+fcon1000Milwaukee_sub98971,F,63.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub98971.stripped.N4.nii.gz
+sub-CC420261,F,54.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420261.stripped.N4.nii.gz
+DLBS0028390_session_1,M,22.5,dlbs_storage/DLBS_bias_corrected/0028390_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub73317,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub73317.stripped.N4.nii.gz
+fcon1000Cambridge_sub67117,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub67117.stripped.N4.nii.gz
+ADNI1457865,F,64.5,adni_storage/ADNI_nii_gz_bias_corrected/I1457865_137_S_6903.stripped.N4.nii.gz
+DLBS0028351_session_1,M,83.73,dlbs_storage/DLBS_bias_corrected/0028351_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub50859,F,79.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub50859.stripped.N4.nii.gz
+ADNI1173877,F,62.2,adni_storage/ADNI_nii_gz_bias_corrected/I1173877_067_S_6525.stripped.N4.nii.gz
+fcon1000Beijing_sub92430,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub92430.stripped.N4.nii.gz
+ADNI952046,F,60.2,adni_storage/ADNI_nii_gz_bias_corrected/I952046_037_S_6141.stripped.N4.nii.gz
+fcon1000Cambridge_sub88853,F,29.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub88853.stripped.N4.nii.gz
+DLBS0028575_session_1,M,81.02,dlbs_storage/DLBS_bias_corrected/0028575_session_1.stripped.N4.nii.gz
+sub-CC220713,F,33.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220713.stripped.N4.nii.gz
+DLBS0028600_session_1,M,70.31,dlbs_storage/DLBS_bias_corrected/0028600_session_1.stripped.N4.nii.gz
+sub-CC110174,F,25.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110174.stripped.N4.nii.gz
+fcon1000Beijing_sub59448,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub59448.stripped.N4.nii.gz
+fcon1000Berlin_sub95068,M,26.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub95068.stripped.N4.nii.gz
+ADNI944155,F,62.2,adni_storage/ADNI_nii_gz_bias_corrected/I944155_116_S_6129.stripped.N4.nii.gz
+sub-CC220419,M,31.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220419.stripped.N4.nii.gz
+fcon1000Beijing_sub17642,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub17642.stripped.N4.nii.gz
+sub-CC520215,F,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520215.stripped.N4.nii.gz
+fcon1000Bangor_sub36736,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub36736.stripped.N4.nii.gz
+DLBS0028480_session_1,F,40.26,dlbs_storage/DLBS_bias_corrected/0028480_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub13093,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub13093.stripped.N4.nii.gz
+fcon1000NewYork_sub69779,M,40.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub69779.stripped.N4.nii.gz
+fcon1000Cambridge_sub12346,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub12346.stripped.N4.nii.gz
+ADNI62032,F,56.8,adni_storage/ADNI_nii_gz_bias_corrected/I62032_016_S_1121.stripped.N4.nii.gz
+sub-CC410169,F,49.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410169.stripped.N4.nii.gz
+sub-CC310450,F,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310450.stripped.N4.nii.gz
+DLBS0028520_session_1,F,65.52,dlbs_storage/DLBS_bias_corrected/0028520_session_1.stripped.N4.nii.gz
+sub-CC510639,M,62.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510639.stripped.N4.nii.gz
+sub-CC420286,M,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420286.stripped.N4.nii.gz
+fcon1000Leiden_sub57187,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub57187.stripped.N4.nii.gz
+sub-CC312222,F,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC312222.stripped.N4.nii.gz
+fcon1000Bangor_sub03557,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub03557.stripped.N4.nii.gz
+DLBS0028545_session_1,F,72.25,dlbs_storage/DLBS_bias_corrected/0028545_session_1.stripped.N4.nii.gz
+fcon1000Leiden_sub38454,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub38454.stripped.N4.nii.gz
+fcon1000Beijing_sub17603,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub17603.stripped.N4.nii.gz
+DLBS0028405_session_1,F,86.45,dlbs_storage/DLBS_bias_corrected/0028405_session_1.stripped.N4.nii.gz
+sub-CC220284,F,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220284.stripped.N4.nii.gz
+DLBS0028629_session_1,F,81.12,dlbs_storage/DLBS_bias_corrected/0028629_session_1.stripped.N4.nii.gz
+sub-CC620454,F,75.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620454.stripped.N4.nii.gz
+ADNI10952529,M,63.9,adni_storage/ADNI_nii_gz_bias_corrected/I10952529_037_S_10062.stripped.N4.nii.gz
+sub-CC420776,F,49.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420776.stripped.N4.nii.gz
+fcon1000Cambridge_sub55114,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub55114.stripped.N4.nii.gz
+sub-CC210526,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210526.stripped.N4.nii.gz
+fcon1000AnnArbor_sub04111,M,25.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub04111.stripped.N4.nii.gz
+fcon1000Beijing_sub45552,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub45552.stripped.N4.nii.gz
+sub-CC710214,M,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710214.stripped.N4.nii.gz
+sub-CC510329,M,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510329.stripped.N4.nii.gz
+DLBS0028454_session_1,M,42.92,dlbs_storage/DLBS_bias_corrected/0028454_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub30421,M,17.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub30421.stripped.N4.nii.gz
+DLBS0028368_session_1,M,79.65,dlbs_storage/DLBS_bias_corrected/0028368_session_1.stripped.N4.nii.gz
+fcon1000NewYork_sub84371,M,38.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub84371.stripped.N4.nii.gz
+ADNI1558064,F,61.8,adni_storage/ADNI_nii_gz_bias_corrected/I1558064_011_S_7048.stripped.N4.nii.gz
+fcon1000Cambridge_sub00156,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub00156.stripped.N4.nii.gz
+sub-CC510255,M,62.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510255.stripped.N4.nii.gz
+sub-CC210304,F,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210304.stripped.N4.nii.gz
+fcon1000NewYork_sub29935,F,16.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub29935.stripped.N4.nii.gz
+fcon1000Beijing_sub73421,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub73421.stripped.N4.nii.gz
+DLBS0028369_session_1,F,62.67,dlbs_storage/DLBS_bias_corrected/0028369_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub20317,M,15.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub20317.stripped.N4.nii.gz
+sub-CC320870,F,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320870.stripped.N4.nii.gz
+fcon1000Cambridge_sub14194,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub14194.stripped.N4.nii.gz
+sub-CC520391,F,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520391.stripped.N4.nii.gz
+sub-CC110319,F,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110319.stripped.N4.nii.gz
+fcon1000AnnArbor_sub72416,M,80.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub72416.stripped.N4.nii.gz
+ADNI263857,F,56.5,adni_storage/ADNI_nii_gz_bias_corrected/I263857_031_S_4024.stripped.N4.nii.gz
+ADNI10888807,M,60.3,adni_storage/ADNI_nii_gz_bias_corrected/I10888807_019_S_10164.stripped.N4.nii.gz
+fcon1000AnnArbor_sub70106,M,17.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub70106.stripped.N4.nii.gz
+DLBS0028558_session_1,F,24.83,dlbs_storage/DLBS_bias_corrected/0028558_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub34781,M,19.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub34781.stripped.N4.nii.gz
+fcon1000Atlanta_sub32093,F,27.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub32093.stripped.N4.nii.gz
+sub-CC520560,M,65.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520560.stripped.N4.nii.gz
+sub-CC621248,F,72.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC621248.stripped.N4.nii.gz
+sub-CC420396,M,53.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420396.stripped.N4.nii.gz
+ADNI1146201,F,61.4,adni_storage/ADNI_nii_gz_bias_corrected/I1146201_037_S_6141.stripped.N4.nii.gz
+sub-CC122172,M,27.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC122172.stripped.N4.nii.gz
+fcon1000Beijing_sub11344,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub11344.stripped.N4.nii.gz
+sub-CC321000,M,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321000.stripped.N4.nii.gz
+sub-CC620026,F,70.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620026.stripped.N4.nii.gz
+sub-CC320500,M,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320500.stripped.N4.nii.gz
+sub-CC721052,M,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721052.stripped.N4.nii.gz
+fcon1000Beijing_sub28907,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub28907.stripped.N4.nii.gz
+fcon1000Cambridge_sub08204,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub08204.stripped.N4.nii.gz
+fcon1000Cambridge_sub16846,M,29.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub16846.stripped.N4.nii.gz
+DLBS0028341_session_1,F,34.81,dlbs_storage/DLBS_bias_corrected/0028341_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub52300,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub52300.stripped.N4.nii.gz
+DLBS0028393_session_1,M,61.69,dlbs_storage/DLBS_bias_corrected/0028393_session_1.stripped.N4.nii.gz
+DLBS0028421_session_1,M,67.05,dlbs_storage/DLBS_bias_corrected/0028421_session_1.stripped.N4.nii.gz
+sub-CC520013,M,67.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520013.stripped.N4.nii.gz
+sub-CC210314,F,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210314.stripped.N4.nii.gz
+fcon1000AnnArbor_sub86367,M,15.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub86367.stripped.N4.nii.gz
+sub-CC420143,F,53.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420143.stripped.N4.nii.gz
+sub-CC320206,M,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320206.stripped.N4.nii.gz
+DLBS0028492_session_1,M,52.89,dlbs_storage/DLBS_bias_corrected/0028492_session_1.stripped.N4.nii.gz
+sub-CC321506,F,39.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321506.stripped.N4.nii.gz
+DLBS0028493_session_1,M,53.36,dlbs_storage/DLBS_bias_corrected/0028493_session_1.stripped.N4.nii.gz
+ADNI1461223,M,61.6,adni_storage/ADNI_nii_gz_bias_corrected/I1461223_070_S_6966.stripped.N4.nii.gz
+fcon1000Beijing_sub30310,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub30310.stripped.N4.nii.gz
+ADNI1142218,F,60.9,adni_storage/ADNI_nii_gz_bias_corrected/I1142218_033_S_6697.stripped.N4.nii.gz
+sub-CC723197,F,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC723197.stripped.N4.nii.gz
+ADNI341787,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I341787_130_S_4982.stripped.N4.nii.gz
+DLBS0028333_session_1,F,76.45,dlbs_storage/DLBS_bias_corrected/0028333_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub95959,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub95959.stripped.N4.nii.gz
+DLBS0028530_session_1,F,49.86,dlbs_storage/DLBS_bias_corrected/0028530_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub22201,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub22201.stripped.N4.nii.gz
+ADNI321225,F,57.8,adni_storage/ADNI_nii_gz_bias_corrected/I321225_041_S_4874.stripped.N4.nii.gz
+ADNI1020137,F,62.6,adni_storage/ADNI_nii_gz_bias_corrected/I1020137_019_S_6483.stripped.N4.nii.gz
+ADNI1553005,M,63.7,adni_storage/ADNI_nii_gz_bias_corrected/I1553005_941_S_7051.stripped.N4.nii.gz
+ADNI297850,F,57.0,adni_storage/ADNI_nii_gz_bias_corrected/I297850_031_S_4024.stripped.N4.nii.gz
+ADNI1591455,F,62.7,adni_storage/ADNI_nii_gz_bias_corrected/I1591455_014_S_7072.stripped.N4.nii.gz
+DLBS0028468_session_1,F,44.31,dlbs_storage/DLBS_bias_corrected/0028468_session_1.stripped.N4.nii.gz
+ADNI400949,M,62.8,adni_storage/ADNI_nii_gz_bias_corrected/I400949_073_S_4795.stripped.N4.nii.gz
+sub-CC510259,F,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510259.stripped.N4.nii.gz
+ADNI1061955,F,58.0,adni_storage/ADNI_nii_gz_bias_corrected/I1061955_109_S_6364.stripped.N4.nii.gz
+fcon1000Cleveland_sub47482,F,28.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub47482.stripped.N4.nii.gz
+sub-CC620005,F,74.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620005.stripped.N4.nii.gz
+DLBS0028413_session_1,F,81.57,dlbs_storage/DLBS_bias_corrected/0028413_session_1.stripped.N4.nii.gz
+DLBS0028415_session_1,F,32.05,dlbs_storage/DLBS_bias_corrected/0028415_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub64969,F,21.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub64969.stripped.N4.nii.gz
+sub-CC510256,M,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510256.stripped.N4.nii.gz
+fcon1000Leiden_sub09796,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub09796.stripped.N4.nii.gz
+sub-CC310410,F,41.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310410.stripped.N4.nii.gz
+ADNI1565078,M,57.6,adni_storage/ADNI_nii_gz_bias_corrected/I1565078_067_S_7059.stripped.N4.nii.gz
+sub-CC310052,F,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310052.stripped.N4.nii.gz
+sub-CC420236,M,53.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420236.stripped.N4.nii.gz
+ADNI363013,F,61.4,adni_storage/ADNI_nii_gz_bias_corrected/I363013_036_S_4894.stripped.N4.nii.gz
+DLBS0028522_session_1,F,34.99,dlbs_storage/DLBS_bias_corrected/0028522_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub91399,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub91399.stripped.N4.nii.gz
+fcon1000NewYork_sub38088,M,16.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub38088.stripped.N4.nii.gz
+DLBS0028348_session_1,M,23.15,dlbs_storage/DLBS_bias_corrected/0028348_session_1.stripped.N4.nii.gz
+fcon1000NewYork_sub48830,F,22.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub48830.stripped.N4.nii.gz
+fcon1000AnnArbor_sub39635,F,19.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub39635.stripped.N4.nii.gz
+fcon1000NewYork_sub31671,M,12.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub31671.stripped.N4.nii.gz
+fcon1000Cambridge_sub76745,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub76745.stripped.N4.nii.gz
+fcon1000Beijing_sub89592,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub89592.stripped.N4.nii.gz
+ADNI387592,F,61.8,adni_storage/ADNI_nii_gz_bias_corrected/I387592_036_S_4894.stripped.N4.nii.gz
+ADNI247840,F,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I247840_037_S_4028.stripped.N4.nii.gz
+fcon1000Beijing_sub18326,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub18326.stripped.N4.nii.gz
+ADNI1634179,F,58.0,adni_storage/ADNI_nii_gz_bias_corrected/I1634179_013_S_7103.stripped.N4.nii.gz
+sub-CC710154,M,83.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710154.stripped.N4.nii.gz
+sub-CC410097,F,49.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410097.stripped.N4.nii.gz
+fcon1000Beijing_sub85543,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub85543.stripped.N4.nii.gz
+ADNI105503,F,63.9,adni_storage/ADNI_nii_gz_bias_corrected/I105503_067_S_1253.stripped.N4.nii.gz
+fcon1000AnnArbor_sub75922,M,18.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub75922.stripped.N4.nii.gz
+sub-CC220151,M,30.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220151.stripped.N4.nii.gz
+fcon1000Cambridge_sub68101,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub68101.stripped.N4.nii.gz
+sub-CC120347,F,21.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120347.stripped.N4.nii.gz
+fcon1000NewYork_sub73035,M,26.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub73035.stripped.N4.nii.gz
+sub-CC620262,F,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620262.stripped.N4.nii.gz
+DLBS0028383_session_1,M,72.96,dlbs_storage/DLBS_bias_corrected/0028383_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub53998,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub53998.stripped.N4.nii.gz
+fcon1000Beijing_sub92859,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub92859.stripped.N4.nii.gz
+ADNI1482729,F,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I1482729_067_S_6529.stripped.N4.nii.gz
+ADNI10951704,M,64.2,adni_storage/ADNI_nii_gz_bias_corrected/I10951704_035_S_10241.stripped.N4.nii.gz
+fcon1000Milwaukee_sub21350,F,61.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub21350.stripped.N4.nii.gz
+DLBS0028463_session_1,F,68.35,dlbs_storage/DLBS_bias_corrected/0028463_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub88464,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub88464.stripped.N4.nii.gz
+fcon1000Cambridge_sub78547,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub78547.stripped.N4.nii.gz
+sub-CC110411,M,25.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110411.stripped.N4.nii.gz
+sub-CC510242,M,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510242.stripped.N4.nii.gz
+ADNI1469825,F,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I1469825_073_S_6929.stripped.N4.nii.gz
+ADNI1028361,M,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I1028361_109_S_6406.stripped.N4.nii.gz
+ADNI346860,F,56.9,adni_storage/ADNI_nii_gz_bias_corrected/I346860_128_S_4607.stripped.N4.nii.gz
+fcon1000Cambridge_sub77989,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub77989.stripped.N4.nii.gz
+ADNI10955657,M,58.4,adni_storage/ADNI_nii_gz_bias_corrected/I10955657_389_S_10156.stripped.N4.nii.gz
+fcon1000Cambridge_sub15905,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub15905.stripped.N4.nii.gz
+sub-CC720103,M,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720103.stripped.N4.nii.gz
+fcon1000NewYork_sub07578,F,20.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub07578.stripped.N4.nii.gz
+sub-CC221648,F,30.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221648.stripped.N4.nii.gz
+ADNI47973,F,62.8,adni_storage/ADNI_nii_gz_bias_corrected/I47973_067_S_1253.stripped.N4.nii.gz
+DLBS0028539_session_1,F,24.22,dlbs_storage/DLBS_bias_corrected/0028539_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub20948,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub20948.stripped.N4.nii.gz
+fcon1000Baltimore_sub37548,M,30.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub37548.stripped.N4.nii.gz
+fcon1000Beijing_sub31729,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub31729.stripped.N4.nii.gz
+sub-CC221755,M,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221755.stripped.N4.nii.gz
+sub-CC620685,F,77.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620685.stripped.N4.nii.gz
+sub-CC121200,F,26.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121200.stripped.N4.nii.gz
+fcon1000Newark_sub32580,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub32580.stripped.N4.nii.gz
+DLBS0028540_session_1,F,71.04,dlbs_storage/DLBS_bias_corrected/0028540_session_1.stripped.N4.nii.gz
+sub-CC110101,M,23.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110101.stripped.N4.nii.gz
+fcon1000NewYork_sub50559,M,22.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub50559.stripped.N4.nii.gz
+DLBS0028589_session_1,F,59.75,dlbs_storage/DLBS_bias_corrected/0028589_session_1.stripped.N4.nii.gz
+fcon1000Baltimore_sub73823,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub73823.stripped.N4.nii.gz
+sub-CC321976,F,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321976.stripped.N4.nii.gz
+fcon1000Munchen_sub72095,F,71.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub72095.stripped.N4.nii.gz
+sub-CC220198,F,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220198.stripped.N4.nii.gz
+ADNI389099,F,59.6,adni_storage/ADNI_nii_gz_bias_corrected/I389099_067_S_5205.stripped.N4.nii.gz
+fcon1000Bangor_sub81464,M,38.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub81464.stripped.N4.nii.gz
+sub-CC221585,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221585.stripped.N4.nii.gz
+fcon1000Cambridge_sub45354,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub45354.stripped.N4.nii.gz
+fcon1000Beijing_sub16943,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub16943.stripped.N4.nii.gz
+ADNI10302840,F,59.5,adni_storage/ADNI_nii_gz_bias_corrected/I10302840_016_S_6773.stripped.N4.nii.gz
+sub-CC620090,F,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620090.stripped.N4.nii.gz
+fcon1000Bangor_sub82625,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub82625.stripped.N4.nii.gz
+ADNI1609469,F,60.8,adni_storage/ADNI_nii_gz_bias_corrected/I1609469_035_S_6927.stripped.N4.nii.gz
+sub-CC320002,M,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320002.stripped.N4.nii.gz
+fcon1000Milwaukee_sub87910,M,47.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub87910.stripped.N4.nii.gz
+DLBS0028499_session_1,M,51.56,dlbs_storage/DLBS_bias_corrected/0028499_session_1.stripped.N4.nii.gz
+sub-CC420202,M,51.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420202.stripped.N4.nii.gz
+fcon1000Cambridge_sub63412,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub63412.stripped.N4.nii.gz
+sub-CC510354,F,62.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510354.stripped.N4.nii.gz
+ADNI10244463,F,61.8,adni_storage/ADNI_nii_gz_bias_corrected/I10244463_035_S_6927.stripped.N4.nii.gz
+sub-CC510551,M,61.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510551.stripped.N4.nii.gz
+fcon1000Leiden_sub99856,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub99856.stripped.N4.nii.gz
+fcon1000AnnArbor_sub11043,F,21.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub11043.stripped.N4.nii.gz
+DLBS0028417_session_1,F,31.77,dlbs_storage/DLBS_bias_corrected/0028417_session_1.stripped.N4.nii.gz
+sub-CC121411,F,26.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121411.stripped.N4.nii.gz
+fcon1000Beijing_sub56136,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub56136.stripped.N4.nii.gz
+DLBS0028435_session_1,M,33.01,dlbs_storage/DLBS_bias_corrected/0028435_session_1.stripped.N4.nii.gz
+DLBS0028366_session_1,F,23.53,dlbs_storage/DLBS_bias_corrected/0028366_session_1.stripped.N4.nii.gz
+ADNI1590221,M,61.4,adni_storage/ADNI_nii_gz_bias_corrected/I1590221_053_S_7086.stripped.N4.nii.gz
+fcon1000Cambridge_sub58470,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub58470.stripped.N4.nii.gz
+fcon1000Milwaukee_sub00917,F,54.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub00917.stripped.N4.nii.gz
+ADNI290869,M,60.0,adni_storage/ADNI_nii_gz_bias_corrected/I290869_128_S_4599.stripped.N4.nii.gz
+ADNI140027,M,62.1,adni_storage/ADNI_nii_gz_bias_corrected/I140027_020_S_1288.stripped.N4.nii.gz
+fcon1000Beijing_sub50498,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub50498.stripped.N4.nii.gz
+DLBS0028380_session_1,F,83.22,dlbs_storage/DLBS_bias_corrected/0028380_session_1.stripped.N4.nii.gz
+DLBS0028636_session_1,F,65.12,dlbs_storage/DLBS_bias_corrected/0028636_session_1.stripped.N4.nii.gz
+fcon1000Munchen_sub70942,M,70.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub70942.stripped.N4.nii.gz
+ADNI1592054,F,57.3,adni_storage/ADNI_nii_gz_bias_corrected/I1592054_014_S_7080.stripped.N4.nii.gz
+DLBS0028378_session_1,F,69.19,dlbs_storage/DLBS_bias_corrected/0028378_session_1.stripped.N4.nii.gz
+fcon1000Leiden_sub12255,M,25.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub12255.stripped.N4.nii.gz
+sub-CC620121,M,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620121.stripped.N4.nii.gz
+sub-CC321087,F,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321087.stripped.N4.nii.gz
+ADNI1539706,F,59.7,adni_storage/ADNI_nii_gz_bias_corrected/I1539706_019_S_6635.stripped.N4.nii.gz
+ADNI73380,F,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I73380_051_S_1072.stripped.N4.nii.gz
+fcon1000Beijing_sub41621,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub41621.stripped.N4.nii.gz
+fcon1000Cambridge_sub78614,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub78614.stripped.N4.nii.gz
+ADNI342681,F,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I342681_073_S_5016.stripped.N4.nii.gz
+fcon1000Beijing_sub82826,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub82826.stripped.N4.nii.gz
+fcon1000Beijing_sub93856,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub93856.stripped.N4.nii.gz
+fcon1000Milwaukee_sub28782,F,64.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub28782.stripped.N4.nii.gz
+sub-CC410101,M,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410101.stripped.N4.nii.gz
+ADNI1583750,F,63.0,adni_storage/ADNI_nii_gz_bias_corrected/I1583750_067_S_7068.stripped.N4.nii.gz
+sub-CC610292,F,72.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610292.stripped.N4.nii.gz
+DLBS0028371_session_1,M,31.11,dlbs_storage/DLBS_bias_corrected/0028371_session_1.stripped.N4.nii.gz
+sub-CC120061,M,19.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120061.stripped.N4.nii.gz
+ADNI10454329,F,56.8,adni_storage/ADNI_nii_gz_bias_corrected/I10454329_003_S_10073.stripped.N4.nii.gz
+fcon1000Cambridge_sub40635,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub40635.stripped.N4.nii.gz
+DLBS0028581_session_1,F,74.58,dlbs_storage/DLBS_bias_corrected/0028581_session_1.stripped.N4.nii.gz
+ADNI302014,F,64.6,adni_storage/ADNI_nii_gz_bias_corrected/I302014_037_S_4028.stripped.N4.nii.gz
+sub-CC321544,M,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321544.stripped.N4.nii.gz
+fcon1000Beijing_sub08251,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub08251.stripped.N4.nii.gz
+DLBS0028448_session_1,M,69.32,dlbs_storage/DLBS_bias_corrected/0028448_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub83430,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub83430.stripped.N4.nii.gz
+fcon1000Cambridge_sub82113,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub82113.stripped.N4.nii.gz
+sub-CC410121,F,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410121.stripped.N4.nii.gz
+DLBS0028498_session_1,M,67.16,dlbs_storage/DLBS_bias_corrected/0028498_session_1.stripped.N4.nii.gz
+sub-CC520597,M,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520597.stripped.N4.nii.gz
+sub-CC420356,M,54.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420356.stripped.N4.nii.gz
+ADNI293371,F,56.3,adni_storage/ADNI_nii_gz_bias_corrected/I293371_128_S_4607.stripped.N4.nii.gz
+DLBS0028339_session_1,F,65.23,dlbs_storage/DLBS_bias_corrected/0028339_session_1.stripped.N4.nii.gz
+sub-CC420729,F,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420729.stripped.N4.nii.gz
+ADNI10369709,F,58.5,adni_storage/ADNI_nii_gz_bias_corrected/I10369709_073_S_10054.stripped.N4.nii.gz
+fcon1000Berlin_sub86111,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub86111.stripped.N4.nii.gz
+ADNI495308,F,57.3,adni_storage/ADNI_nii_gz_bias_corrected/I495308_128_S_4607.stripped.N4.nii.gz
+ADNI321230,F,57.8,adni_storage/ADNI_nii_gz_bias_corrected/I321230_041_S_4874.stripped.N4.nii.gz
+ADNI461101,F,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I461101_137_S_0994.stripped.N4.nii.gz
+DLBS0028403_session_1,M,68.71,dlbs_storage/DLBS_bias_corrected/0028403_session_1.stripped.N4.nii.gz
+fcon1000Leiden_sub30943,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub30943.stripped.N4.nii.gz
+fcon1000Baltimore_sub94042,F,27.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub94042.stripped.N4.nii.gz
+fcon1000Beijing_sub40427,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub40427.stripped.N4.nii.gz
+sub-CC721392,M,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721392.stripped.N4.nii.gz
+sub-CC520552,M,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520552.stripped.N4.nii.gz
+ADNI256321,M,56.3,adni_storage/ADNI_nii_gz_bias_corrected/I256321_014_S_4039.stripped.N4.nii.gz
+ADNI10568658,M,56.7,adni_storage/ADNI_nii_gz_bias_corrected/I10568658_941_S_10065.stripped.N4.nii.gz
+fcon1000Leiden_sub36743,M,25.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub36743.stripped.N4.nii.gz
+ADNI1498335,F,57.3,adni_storage/ADNI_nii_gz_bias_corrected/I1498335_016_S_6773.stripped.N4.nii.gz
+fcon1000AnnArbor_sub13636,M,29.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub13636.stripped.N4.nii.gz
+fcon1000AnnArbor_sub49687,M,13.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub49687.stripped.N4.nii.gz
+sub-CC320814,F,43.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320814.stripped.N4.nii.gz
+ADNI1446680,F,59.8,adni_storage/ADNI_nii_gz_bias_corrected/I1446680_070_S_6942.stripped.N4.nii.gz
+sub-CC620044,F,73.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620044.stripped.N4.nii.gz
+sub-CC510220,M,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510220.stripped.N4.nii.gz
+ADNI313456,M,60.3,adni_storage/ADNI_nii_gz_bias_corrected/I313456_128_S_4599.stripped.N4.nii.gz
+ADNI10970157,M,56.0,adni_storage/ADNI_nii_gz_bias_corrected/I10970157_073_S_10254.stripped.N4.nii.gz
+sub-CC412021,F,54.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC412021.stripped.N4.nii.gz
+sub-CC610212,F,77.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610212.stripped.N4.nii.gz
+sub-CC220843,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220843.stripped.N4.nii.gz
+sub-CC620610,M,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620610.stripped.N4.nii.gz
+ADNI1558652,F,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I1558652_014_S_6076.stripped.N4.nii.gz
+fcon1000Cambridge_sub65682,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub65682.stripped.N4.nii.gz
+DLBS0028385_session_1,F,61.53,dlbs_storage/DLBS_bias_corrected/0028385_session_1.stripped.N4.nii.gz
+sub-CC410432,M,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410432.stripped.N4.nii.gz
+sub-CC321529,F,39.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321529.stripped.N4.nii.gz
+sub-CC120640,F,26.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120640.stripped.N4.nii.gz
+ADNI460665,M,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I460665_073_S_4795.stripped.N4.nii.gz
+fcon1000AnnArbor_sub64831,M,19.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub64831.stripped.N4.nii.gz
+ADNI10349528,F,58.8,adni_storage/ADNI_nii_gz_bias_corrected/I10349528_035_S_7004.stripped.N4.nii.gz
+sub-CC221580,F,31.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221580.stripped.N4.nii.gz
+ADNI254253,M,62.3,adni_storage/ADNI_nii_gz_bias_corrected/I254253_116_S_4195.stripped.N4.nii.gz
+sub-CC410323,F,51.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410323.stripped.N4.nii.gz
+DLBS0028464_session_1,F,82.43,dlbs_storage/DLBS_bias_corrected/0028464_session_1.stripped.N4.nii.gz
+fcon1000Milwaukee_sub14692,F,53.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub14692.stripped.N4.nii.gz
+fcon1000Cambridge_sub82213,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub82213.stripped.N4.nii.gz
+fcon1000Bangor_sub46870,M,28.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub46870.stripped.N4.nii.gz
+sub-CC220203,F,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220203.stripped.N4.nii.gz
+fcon1000Cleveland_sub85091,F,52.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub85091.stripped.N4.nii.gz
+sub-CC310051,F,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310051.stripped.N4.nii.gz
+sub-CC120182,M,26.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120182.stripped.N4.nii.gz
+ADNI335130,M,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I335130_073_S_4552.stripped.N4.nii.gz
+sub-CC510483,F,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510483.stripped.N4.nii.gz
+fcon1000NewYork_sub13384,F,28.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub13384.stripped.N4.nii.gz
+DLBS0028327_session_1,F,73.29,dlbs_storage/DLBS_bias_corrected/0028327_session_1.stripped.N4.nii.gz
+ADNI10666738,F,59.9,adni_storage/ADNI_nii_gz_bias_corrected/I10666738_011_S_10045.stripped.N4.nii.gz
+DLBS0028580_session_1,F,60.28,dlbs_storage/DLBS_bias_corrected/0028580_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub86115,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub86115.stripped.N4.nii.gz
+sub-CC410040,M,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410040.stripped.N4.nii.gz
+ADNI368463,M,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I368463_073_S_4552.stripped.N4.nii.gz
+sub-CC520042,M,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520042.stripped.N4.nii.gz
+fcon1000Cambridge_sub53615,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub53615.stripped.N4.nii.gz
+DLBS0028507_session_1,M,36.91,dlbs_storage/DLBS_bias_corrected/0028507_session_1.stripped.N4.nii.gz
+sub-CC510434,F,65.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510434.stripped.N4.nii.gz
+fcon1000Beijing_sub82980,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub82980.stripped.N4.nii.gz
+fcon1000NewYork_sub30247,F,24.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub30247.stripped.N4.nii.gz
+ADNI37606,F,60.1,adni_storage/ADNI_nii_gz_bias_corrected/I37606_051_S_1072.stripped.N4.nii.gz
+DLBS0028562_session_1,F,24.29,dlbs_storage/DLBS_bias_corrected/0028562_session_1.stripped.N4.nii.gz
+DLBS0028543_session_1,F,81.28,dlbs_storage/DLBS_bias_corrected/0028543_session_1.stripped.N4.nii.gz
+fcon1000Cleveland_sub22736,F,28.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub22736.stripped.N4.nii.gz
+sub-CC420198,M,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420198.stripped.N4.nii.gz
+sub-CC410248,M,54.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410248.stripped.N4.nii.gz
+sub-CC121317,F,25.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121317.stripped.N4.nii.gz
+ADNI417674,M,62.0,adni_storage/ADNI_nii_gz_bias_corrected/I417674_128_S_4599.stripped.N4.nii.gz
+sub-CC220806,F,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220806.stripped.N4.nii.gz
+DLBS0028419_session_1,F,78.87,dlbs_storage/DLBS_bias_corrected/0028419_session_1.stripped.N4.nii.gz
+DLBS0028356_session_1,F,43.21,dlbs_storage/DLBS_bias_corrected/0028356_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub40037,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub40037.stripped.N4.nii.gz
+fcon1000Beijing_sub32517,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub32517.stripped.N4.nii.gz
+ADNI1118422,F,59.3,adni_storage/ADNI_nii_gz_bias_corrected/I1118422_006_S_6651.stripped.N4.nii.gz
+sub-CC221935,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221935.stripped.N4.nii.gz
+ADNI958916,F,55.6,adni_storage/ADNI_nii_gz_bias_corrected/I958916_070_S_6191.stripped.N4.nii.gz
+sub-CC420182,F,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420182.stripped.N4.nii.gz
+ADNI335038,M,63.4,adni_storage/ADNI_nii_gz_bias_corrected/I335038_116_S_4195.stripped.N4.nii.gz
+sub-CC520480,M,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520480.stripped.N4.nii.gz
+fcon1000Berlin_sub85681,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub85681.stripped.N4.nii.gz
+fcon1000Cambridge_sub07798,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub07798.stripped.N4.nii.gz
+fcon1000NewYork_sub02503,M,9.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub02503.stripped.N4.nii.gz
+ADNI1076708,F,55.2,adni_storage/ADNI_nii_gz_bias_corrected/I1076708_037_S_6620.stripped.N4.nii.gz
+sub-CC121194,F,24.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121194.stripped.N4.nii.gz
+sub-CC420148,F,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420148.stripped.N4.nii.gz
+sub-CC220535,M,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220535.stripped.N4.nii.gz
+sub-CC110033,M,24.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110033.stripped.N4.nii.gz
+fcon1000Beijing_sub17586,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub17586.stripped.N4.nii.gz
+ADNI1043769,F,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I1043769_003_S_6490.stripped.N4.nii.gz
+sub-CC223115,M,36.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC223115.stripped.N4.nii.gz
+sub-CC510039,F,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510039.stripped.N4.nii.gz
+DLBS0028598_session_1,F,84.06,dlbs_storage/DLBS_bias_corrected/0028598_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub29044,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub29044.stripped.N4.nii.gz
+DLBS0028606_session_1,F,81.91,dlbs_storage/DLBS_bias_corrected/0028606_session_1.stripped.N4.nii.gz
+fcon1000NewYork_sub28808,F,23.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub28808.stripped.N4.nii.gz
+DLBS0028624_session_1,F,81.96,dlbs_storage/DLBS_bias_corrected/0028624_session_1.stripped.N4.nii.gz
+sub-CC510321,F,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510321.stripped.N4.nii.gz
+ADNI358814,F,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I358814_019_S_5019.stripped.N4.nii.gz
+ADNI248875,F,56.3,adni_storage/ADNI_nii_gz_bias_corrected/I248875_031_S_4024.stripped.N4.nii.gz
+sub-CC420259,F,51.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420259.stripped.N4.nii.gz
+ADNI1007384,F,56.2,adni_storage/ADNI_nii_gz_bias_corrected/I1007384_035_S_6380.stripped.N4.nii.gz
+DLBS0028536_session_1,M,82.57,dlbs_storage/DLBS_bias_corrected/0028536_session_1.stripped.N4.nii.gz
+DLBS0028365_session_1,F,22.82,dlbs_storage/DLBS_bias_corrected/0028365_session_1.stripped.N4.nii.gz
+sub-CC610052,M,77.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610052.stripped.N4.nii.gz
+fcon1000Beijing_sub14238,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub14238.stripped.N4.nii.gz
+fcon1000Beijing_sub48676,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub48676.stripped.N4.nii.gz
+sub-CC220610,F,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220610.stripped.N4.nii.gz
+sub-CC610308,M,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610308.stripped.N4.nii.gz
+ADNI10236670,F,61.5,adni_storage/ADNI_nii_gz_bias_corrected/I10236670_168_S_6085.stripped.N4.nii.gz
+fcon1000Milwaukee_sub39259,F,53.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub39259.stripped.N4.nii.gz
+sub-CC520585,F,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520585.stripped.N4.nii.gz
+DLBS0028512_session_1,F,51.64,dlbs_storage/DLBS_bias_corrected/0028512_session_1.stripped.N4.nii.gz
+ADNI944156,F,62.2,adni_storage/ADNI_nii_gz_bias_corrected/I944156_116_S_6129.stripped.N4.nii.gz
+ADNI168100,F,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I168100_051_S_1072.stripped.N4.nii.gz
+ADNI442430,F,59.8,adni_storage/ADNI_nii_gz_bias_corrected/I442430_041_S_4874.stripped.N4.nii.gz
+fcon1000Cambridge_sub53059,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub53059.stripped.N4.nii.gz
+sub-CC420383,F,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420383.stripped.N4.nii.gz
+sub-CC321504,M,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321504.stripped.N4.nii.gz
+fcon1000Beijing_sub41170,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub41170.stripped.N4.nii.gz
+DLBS0028640_session_1,F,33.41,dlbs_storage/DLBS_bias_corrected/0028640_session_1.stripped.N4.nii.gz
+ADNI1399348,F,63.4,adni_storage/ADNI_nii_gz_bias_corrected/I1399348_305_S_6845.stripped.N4.nii.gz
+fcon1000Beijing_sub56757,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub56757.stripped.N4.nii.gz
+ADNI105505,F,63.9,adni_storage/ADNI_nii_gz_bias_corrected/I105505_067_S_1253.stripped.N4.nii.gz
+fcon1000Cambridge_sub07413,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub07413.stripped.N4.nii.gz
+sub-CC610028,M,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610028.stripped.N4.nii.gz
+DLBS0028563_session_1,F,44.97,dlbs_storage/DLBS_bias_corrected/0028563_session_1.stripped.N4.nii.gz
+sub-CC321585,F,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321585.stripped.N4.nii.gz
+fcon1000Atlanta_sub59806,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub59806.stripped.N4.nii.gz
+sub-CC221527,F,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221527.stripped.N4.nii.gz
+ADNI1573570,F,63.1,adni_storage/ADNI_nii_gz_bias_corrected/I1573570_131_S_6805.stripped.N4.nii.gz
+fcon1000AnnArbor_sub56686,F,20.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub56686.stripped.N4.nii.gz
+ADNI10936993,F,55.7,adni_storage/ADNI_nii_gz_bias_corrected/I10936993_033_S_10215.stripped.N4.nii.gz
+fcon1000Cleveland_sub19005,M,29.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub19005.stripped.N4.nii.gz
+fcon1000Atlanta_sub76280,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub76280.stripped.N4.nii.gz
+sub-CC120409,M,18.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120409.stripped.N4.nii.gz
+fcon1000Beijing_sub22595,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub22595.stripped.N4.nii.gz
+fcon1000Leiden_sub13537,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub13537.stripped.N4.nii.gz
+ADNI1344108,F,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I1344108_123_S_6888.stripped.N4.nii.gz
+sub-CC210148,F,30.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210148.stripped.N4.nii.gz
+fcon1000NewYork_sub20732,M,11.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub20732.stripped.N4.nii.gz
+fcon1000Cambridge_sub04187,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub04187.stripped.N4.nii.gz
+ADNI912447,F,63.1,adni_storage/ADNI_nii_gz_bias_corrected/I912447_003_S_6067.stripped.N4.nii.gz
+sub-CC710551,M,85.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710551.stripped.N4.nii.gz
+fcon1000Cambridge_sub93488,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub93488.stripped.N4.nii.gz
+fcon1000Beijing_sub61961,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub61961.stripped.N4.nii.gz
+ADNI280814,F,63.9,adni_storage/ADNI_nii_gz_bias_corrected/I280814_023_S_4448.stripped.N4.nii.gz
+sub-CC321154,M,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321154.stripped.N4.nii.gz
+sub-CC410297,F,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410297.stripped.N4.nii.gz
+sub-CC221565,F,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221565.stripped.N4.nii.gz
+sub-CC510480,M,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510480.stripped.N4.nii.gz
+fcon1000AnnArbor_sub57196,F,77.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub57196.stripped.N4.nii.gz
+ADNI313455,M,60.3,adni_storage/ADNI_nii_gz_bias_corrected/I313455_128_S_4599.stripped.N4.nii.gz
+DLBS0028610_session_1,M,80.82,dlbs_storage/DLBS_bias_corrected/0028610_session_1.stripped.N4.nii.gz
+DLBS0028579_session_1,F,21.77,dlbs_storage/DLBS_bias_corrected/0028579_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub30250,F,19.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub30250.stripped.N4.nii.gz
+fcon1000Beijing_sub52044,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub52044.stripped.N4.nii.gz
+sub-CC620518,M,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620518.stripped.N4.nii.gz
+ADNI1330894,F,62.0,adni_storage/ADNI_nii_gz_bias_corrected/I1330894_305_S_6881.stripped.N4.nii.gz
+fcon1000Leiden_sub58194,M,27.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub58194.stripped.N4.nii.gz
+DLBS0028467_session_1,F,34.4,dlbs_storage/DLBS_bias_corrected/0028467_session_1.stripped.N4.nii.gz
+sub-CC620560,M,72.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620560.stripped.N4.nii.gz
+ADNI1008024,F,59.5,adni_storage/ADNI_nii_gz_bias_corrected/I1008024_177_S_6408.stripped.N4.nii.gz
+ADNI1558656,F,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I1558656_014_S_6076.stripped.N4.nii.gz
+sub-CC710462,F,82.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710462.stripped.N4.nii.gz
+DLBS0028525_session_1,M,60.74,dlbs_storage/DLBS_bias_corrected/0028525_session_1.stripped.N4.nii.gz
+sub-CC721894,F,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721894.stripped.N4.nii.gz
+fcon1000Beijing_sub81074,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub81074.stripped.N4.nii.gz
+sub-CC121397,M,27.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121397.stripped.N4.nii.gz
+ADNI1529914,F,59.4,adni_storage/ADNI_nii_gz_bias_corrected/I1529914_067_S_7033.stripped.N4.nii.gz
+fcon1000Cambridge_sub02591,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub02591.stripped.N4.nii.gz
+fcon1000Baltimore_sub19738,F,27.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub19738.stripped.N4.nii.gz
+fcon1000Cleveland_sub50092,M,56.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub50092.stripped.N4.nii.gz
+sub-CC610101,M,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610101.stripped.N4.nii.gz
+ADNI378273,M,58.1,adni_storage/ADNI_nii_gz_bias_corrected/I378273_014_S_4039.stripped.N4.nii.gz
+ADNI358820,F,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I358820_019_S_5019.stripped.N4.nii.gz
+fcon1000NewYork_sub47087,M,25.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub47087.stripped.N4.nii.gz
+sub-CC620720,F,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620720.stripped.N4.nii.gz
+sub-CC321464,M,45.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321464.stripped.N4.nii.gz
+sub-CC721224,F,87.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721224.stripped.N4.nii.gz
+ADNI1142379,F,61.5,adni_storage/ADNI_nii_gz_bias_corrected/I1142379_022_S_6013.stripped.N4.nii.gz
+ADNI10461527,F,60.0,adni_storage/ADNI_nii_gz_bias_corrected/I10461527_022_S_10040.stripped.N4.nii.gz
+ADNI270531,M,56.5,adni_storage/ADNI_nii_gz_bias_corrected/I270531_014_S_4039.stripped.N4.nii.gz
+fcon1000Berlin_sub18913,F,29.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub18913.stripped.N4.nii.gz
+ADNI65946,M,62.7,adni_storage/ADNI_nii_gz_bias_corrected/I65946_067_S_1185.stripped.N4.nii.gz
+fcon1000Bangor_sub68050,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub68050.stripped.N4.nii.gz
+ADNI1609767,F,61.1,adni_storage/ADNI_nii_gz_bias_corrected/I1609767_073_S_7082.stripped.N4.nii.gz
+fcon1000Cambridge_sub98528,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub98528.stripped.N4.nii.gz
+fcon1000NewYork_sub05208,F,23.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub05208.stripped.N4.nii.gz
+DLBS0028560_session_1,F,57.03,dlbs_storage/DLBS_bias_corrected/0028560_session_1.stripped.N4.nii.gz
+DLBS0028394_session_1,M,20.74,dlbs_storage/DLBS_bias_corrected/0028394_session_1.stripped.N4.nii.gz
+ADNI1327480,F,60.4,adni_storage/ADNI_nii_gz_bias_corrected/I1327480_177_S_6409.stripped.N4.nii.gz
+fcon1000NewYork_sub30860,F,35.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub30860.stripped.N4.nii.gz
+ADNI1516245,F,58.2,adni_storage/ADNI_nii_gz_bias_corrected/I1516245_082_S_6629.stripped.N4.nii.gz
+ADNI393944,F,63.6,adni_storage/ADNI_nii_gz_bias_corrected/I393944_073_S_5023.stripped.N4.nii.gz
+ADNI1493847,M,61.7,adni_storage/ADNI_nii_gz_bias_corrected/I1493847_016_S_7002.stripped.N4.nii.gz
+sub-CC722216,F,86.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC722216.stripped.N4.nii.gz
+fcon1000Cleveland_sub13495,F,45.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub13495.stripped.N4.nii.gz
+fcon1000Beijing_sub52259,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub52259.stripped.N4.nii.gz
+ADNI1183115,M,62.2,adni_storage/ADNI_nii_gz_bias_corrected/I1183115_305_S_6744.stripped.N4.nii.gz
+ADNI393941,F,63.6,adni_storage/ADNI_nii_gz_bias_corrected/I393941_073_S_5023.stripped.N4.nii.gz
+sub-CC110126,F,22.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110126.stripped.N4.nii.gz
+ADNI248872,F,56.3,adni_storage/ADNI_nii_gz_bias_corrected/I248872_031_S_4024.stripped.N4.nii.gz
+DLBS0028373_session_1,F,48.4,dlbs_storage/DLBS_bias_corrected/0028373_session_1.stripped.N4.nii.gz
+ADNI1149157,F,62.8,adni_storage/ADNI_nii_gz_bias_corrected/I1149157_301_S_6698.stripped.N4.nii.gz
+DLBS0028494_session_1,M,74.24,dlbs_storage/DLBS_bias_corrected/0028494_session_1.stripped.N4.nii.gz
+fcon1000Baltimore_sub80221,M,26.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub80221.stripped.N4.nii.gz
+fcon1000AnnArbor_sub18546,M,20.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub18546.stripped.N4.nii.gz
+sub-CC520175,M,61.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520175.stripped.N4.nii.gz
+ADNI950885,M,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I950885_035_S_6160.stripped.N4.nii.gz
+fcon1000Milwaukee_sub56582,F,57.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub56582.stripped.N4.nii.gz
+DLBS0028590_session_1,F,70.79,dlbs_storage/DLBS_bias_corrected/0028590_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub01361,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub01361.stripped.N4.nii.gz
+DLBS0028583_session_1,F,80.54,dlbs_storage/DLBS_bias_corrected/0028583_session_1.stripped.N4.nii.gz
+ADNI344065,F,58.0,adni_storage/ADNI_nii_gz_bias_corrected/I344065_041_S_4874.stripped.N4.nii.gz
+DLBS0028542_session_1,F,69.59,dlbs_storage/DLBS_bias_corrected/0028542_session_1.stripped.N4.nii.gz
+sub-CC320904,M,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320904.stripped.N4.nii.gz
+sub-CC110045,F,24.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110045.stripped.N4.nii.gz
+DLBS0028457_session_1,M,75.67,dlbs_storage/DLBS_bias_corrected/0028457_session_1.stripped.N4.nii.gz
+DLBS0028627_session_1,F,47.3,dlbs_storage/DLBS_bias_corrected/0028627_session_1.stripped.N4.nii.gz
+fcon1000Baltimore_sub54257,F,29.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub54257.stripped.N4.nii.gz
+ADNI884453,F,62.7,adni_storage/ADNI_nii_gz_bias_corrected/I884453_041_S_4874.stripped.N4.nii.gz
+ADNI460668,M,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I460668_073_S_4795.stripped.N4.nii.gz
+fcon1000Beijing_sub17315,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub17315.stripped.N4.nii.gz
+fcon1000Cambridge_sub06037,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub06037.stripped.N4.nii.gz
+fcon1000Cambridge_sub16122,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub16122.stripped.N4.nii.gz
+fcon1000Milwaukee_sub17987,M,47.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub17987.stripped.N4.nii.gz
+sub-CC220132,M,31.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220132.stripped.N4.nii.gz
+ADNI336502,M,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I336502_135_S_4954.stripped.N4.nii.gz
+sub-CC122620,F,24.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC122620.stripped.N4.nii.gz
+sub-CC520398,F,59.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520398.stripped.N4.nii.gz
+ADNI1446304,M,62.2,adni_storage/ADNI_nii_gz_bias_corrected/I1446304_016_S_6941.stripped.N4.nii.gz
+sub-CC210182,F,31.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210182.stripped.N4.nii.gz
+fcon1000Cleveland_sub76139,F,41.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub76139.stripped.N4.nii.gz
+fcon1000Cleveland_sub07835,M,43.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub07835.stripped.N4.nii.gz
+sub-CC310391,M,41.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310391.stripped.N4.nii.gz
+sub-CC510163,F,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510163.stripped.N4.nii.gz
+DLBS0028572_session_1,M,75.37,dlbs_storage/DLBS_bias_corrected/0028572_session_1.stripped.N4.nii.gz
+ADNI359044,M,58.9,adni_storage/ADNI_nii_gz_bias_corrected/I359044_073_S_5090.stripped.N4.nii.gz
+fcon1000Cambridge_sub16390,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub16390.stripped.N4.nii.gz
+sub-CC120319,F,27.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120319.stripped.N4.nii.gz
+DLBS0028491_session_1,F,24.2,dlbs_storage/DLBS_bias_corrected/0028491_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub08816,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub08816.stripped.N4.nii.gz
+ADNI1278681,F,61.7,adni_storage/ADNI_nii_gz_bias_corrected/I1278681_067_S_6529.stripped.N4.nii.gz
+sub-CC221054,M,31.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221054.stripped.N4.nii.gz
+ADNI1017004,M,59.0,adni_storage/ADNI_nii_gz_bias_corrected/I1017004_116_S_6428.stripped.N4.nii.gz
+fcon1000Cambridge_sub84845,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub84845.stripped.N4.nii.gz
+sub-CC620659,F,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620659.stripped.N4.nii.gz
+fcon1000Atlanta_sub07145,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub07145.stripped.N4.nii.gz
+ADNI10435136,M,62.7,adni_storage/ADNI_nii_gz_bias_corrected/I10435136_007_S_10075.stripped.N4.nii.gz
+fcon1000Beijing_sub83728,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub83728.stripped.N4.nii.gz
+fcon1000NewYork_sub29353,F,21.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub29353.stripped.N4.nii.gz
+ADNI388747,M,64.3,adni_storage/ADNI_nii_gz_bias_corrected/I388747_116_S_4195.stripped.N4.nii.gz
+sub-CC221886,M,33.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221886.stripped.N4.nii.gz
+ADNI844180,F,60.4,adni_storage/ADNI_nii_gz_bias_corrected/I844180_022_S_6013.stripped.N4.nii.gz
+ADNI1514928,F,57.9,adni_storage/ADNI_nii_gz_bias_corrected/I1514928_037_S_7006.stripped.N4.nii.gz
+DLBS0028410_session_1,F,74.11,dlbs_storage/DLBS_bias_corrected/0028410_session_1.stripped.N4.nii.gz
+ADNI1516258,F,58.2,adni_storage/ADNI_nii_gz_bias_corrected/I1516258_082_S_6629.stripped.N4.nii.gz
+fcon1000Berlin_sub49134,M,44.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub49134.stripped.N4.nii.gz
+DLBS0028516_session_1,M,88.14,dlbs_storage/DLBS_bias_corrected/0028516_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub92602,M,26.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub92602.stripped.N4.nii.gz
+ADNI274154,M,63.0,adni_storage/ADNI_nii_gz_bias_corrected/I274154_002_S_1155.stripped.N4.nii.gz
+sub-CC520624,F,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520624.stripped.N4.nii.gz
+ADNI1285703,F,59.6,adni_storage/ADNI_nii_gz_bias_corrected/I1285703_131_S_6170.stripped.N4.nii.gz
+fcon1000Atlanta_sub00368,F,28.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub00368.stripped.N4.nii.gz
+fcon1000Cambridge_sub18449,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub18449.stripped.N4.nii.gz
+ADNI1660846,F,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I1660846_019_S_6635.stripped.N4.nii.gz
+sub-CC220901,F,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220901.stripped.N4.nii.gz
+fcon1000Milwaukee_sub76042,F,58.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub76042.stripped.N4.nii.gz
+fcon1000Munchen_sub96591,M,73.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub96591.stripped.N4.nii.gz
+DLBS0028432_session_1,F,53.75,dlbs_storage/DLBS_bias_corrected/0028432_session_1.stripped.N4.nii.gz
+DLBS0028605_session_1,M,63.69,dlbs_storage/DLBS_bias_corrected/0028605_session_1.stripped.N4.nii.gz
+sub-CC520239,F,65.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520239.stripped.N4.nii.gz
+sub-CC120764,M,27.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120764.stripped.N4.nii.gz
+sub-CC620429,M,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620429.stripped.N4.nii.gz
+ADNI132476,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I132476_016_S_1121.stripped.N4.nii.gz
+sub-CC220518,M,30.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220518.stripped.N4.nii.gz
+sub-CC410182,F,53.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410182.stripped.N4.nii.gz
+ADNI365429,F,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I365429_073_S_5023.stripped.N4.nii.gz
+DLBS0028527_session_1,M,23.64,dlbs_storage/DLBS_bias_corrected/0028527_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub05306,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub05306.stripped.N4.nii.gz
+sub-CC710429,M,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710429.stripped.N4.nii.gz
+fcon1000Cambridge_sub20389,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub20389.stripped.N4.nii.gz
+sub-CC120065,F,25.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120065.stripped.N4.nii.gz
+DLBS0028534_session_1,M,54.3,dlbs_storage/DLBS_bias_corrected/0028534_session_1.stripped.N4.nii.gz
+sub-CC710548,M,83.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710548.stripped.N4.nii.gz
+fcon1000Beijing_sub67435,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub67435.stripped.N4.nii.gz
+DLBS0028485_session_1,F,57.23,dlbs_storage/DLBS_bias_corrected/0028485_session_1.stripped.N4.nii.gz
+sub-CC420157,M,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420157.stripped.N4.nii.gz
+sub-CC320698,F,45.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320698.stripped.N4.nii.gz
+ADNI297848,F,57.0,adni_storage/ADNI_nii_gz_bias_corrected/I297848_031_S_4024.stripped.N4.nii.gz
+ADNI415211,F,64.0,adni_storage/ADNI_nii_gz_bias_corrected/I415211_019_S_5019.stripped.N4.nii.gz
+fcon1000Beijing_sub91145,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub91145.stripped.N4.nii.gz
+fcon1000Baltimore_sub23927,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub23927.stripped.N4.nii.gz
+sub-CC110087,F,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110087.stripped.N4.nii.gz
+fcon1000Milwaukee_sub77073,F,52.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub77073.stripped.N4.nii.gz
+sub-CC710099,F,85.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710099.stripped.N4.nii.gz
+DLBS0028619_session_1,M,36.41,dlbs_storage/DLBS_bias_corrected/0028619_session_1.stripped.N4.nii.gz
+ADNI20681,M,64.5,adni_storage/ADNI_nii_gz_bias_corrected/I20681_131_S_0691.stripped.N4.nii.gz
+fcon1000Beijing_sub97442,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub97442.stripped.N4.nii.gz
+DLBS0028528_session_1,M,53.43,dlbs_storage/DLBS_bias_corrected/0028528_session_1.stripped.N4.nii.gz
+ADNI1526394,F,61.4,adni_storage/ADNI_nii_gz_bias_corrected/I1526394_131_S_6170.stripped.N4.nii.gz
+DLBS0028426_session_1,F,82.07,dlbs_storage/DLBS_bias_corrected/0028426_session_1.stripped.N4.nii.gz
+DLBS0028406_session_1,M,47.65,dlbs_storage/DLBS_bias_corrected/0028406_session_1.stripped.N4.nii.gz
+ADNI393947,F,63.6,adni_storage/ADNI_nii_gz_bias_corrected/I393947_073_S_5023.stripped.N4.nii.gz
+sub-CC310385,M,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310385.stripped.N4.nii.gz
+ADNI10932457,M,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I10932457_035_S_7121.stripped.N4.nii.gz
+DLBS0028521_session_1,F,39.74,dlbs_storage/DLBS_bias_corrected/0028521_session_1.stripped.N4.nii.gz
+ADNI1046066,F,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I1046066_041_S_4874.stripped.N4.nii.gz
+fcon1000Milwaukee_sub46312,F,51.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub46312.stripped.N4.nii.gz
+fcon1000Beijing_sub29590,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub29590.stripped.N4.nii.gz
+fcon1000AnnArbor_sub46727,F,13.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub46727.stripped.N4.nii.gz
+sub-CC420888,F,50.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420888.stripped.N4.nii.gz
+sub-CC220234,M,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220234.stripped.N4.nii.gz
+fcon1000Beijing_sub49782,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub49782.stripped.N4.nii.gz
+fcon1000Milwaukee_sub63196,F,49.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub63196.stripped.N4.nii.gz
+sub-CC222956,F,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222956.stripped.N4.nii.gz
+sub-CC610210,M,75.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610210.stripped.N4.nii.gz
+DLBS0028398_session_1,M,83.33,dlbs_storage/DLBS_bias_corrected/0028398_session_1.stripped.N4.nii.gz
+ADNI1151456,M,64.3,adni_storage/ADNI_nii_gz_bias_corrected/I1151456_002_S_6680.stripped.N4.nii.gz
+fcon1000Cambridge_sub90699,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub90699.stripped.N4.nii.gz
+fcon1000Beijing_sub21115,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub21115.stripped.N4.nii.gz
+sub-CC310086,F,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310086.stripped.N4.nii.gz
+sub-CC420173,F,49.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420173.stripped.N4.nii.gz
+DLBS0028571_session_1,M,20.57,dlbs_storage/DLBS_bias_corrected/0028571_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub00306,F,68.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub00306.stripped.N4.nii.gz
+sub-CC222326,M,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222326.stripped.N4.nii.gz
+sub-CC620118,F,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620118.stripped.N4.nii.gz
+ADNI1061954,F,58.0,adni_storage/ADNI_nii_gz_bias_corrected/I1061954_109_S_6364.stripped.N4.nii.gz
+fcon1000Cambridge_sub41773,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub41773.stripped.N4.nii.gz
+sub-CC410390,M,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410390.stripped.N4.nii.gz
+sub-CC610040,M,72.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610040.stripped.N4.nii.gz
+ADNI986502,F,58.4,adni_storage/ADNI_nii_gz_bias_corrected/I986502_094_S_6275.stripped.N4.nii.gz
+fcon1000Leiden_sub04484,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub04484.stripped.N4.nii.gz
+fcon1000Atlanta_sub53122,M,28.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub53122.stripped.N4.nii.gz
+sub-CC321069,F,41.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321069.stripped.N4.nii.gz
+fcon1000Cambridge_sub50454,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub50454.stripped.N4.nii.gz
+ADNI378271,M,58.1,adni_storage/ADNI_nii_gz_bias_corrected/I378271_014_S_4039.stripped.N4.nii.gz
+DLBS0028423_session_1,M,30.69,dlbs_storage/DLBS_bias_corrected/0028423_session_1.stripped.N4.nii.gz
+ADNI10461464,F,56.2,adni_storage/ADNI_nii_gz_bias_corrected/I10461464_033_S_10092.stripped.N4.nii.gz
+fcon1000Beijing_sub51015,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub51015.stripped.N4.nii.gz
+fcon1000Cambridge_sub93269,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub93269.stripped.N4.nii.gz
+fcon1000Beijing_sub50873,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub50873.stripped.N4.nii.gz
+fcon1000Cambridge_sub18295,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub18295.stripped.N4.nii.gz
+ADNI77626,M,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I77626_020_S_1288.stripped.N4.nii.gz
+fcon1000AnnArbor_sub26099,M,73.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub26099.stripped.N4.nii.gz
+DLBS0028430_session_1,F,89.11,dlbs_storage/DLBS_bias_corrected/0028430_session_1.stripped.N4.nii.gz
+sub-CC721291,F,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721291.stripped.N4.nii.gz
+fcon1000Cleveland_sub22935,F,39.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub22935.stripped.N4.nii.gz
+ADNI387580,F,61.8,adni_storage/ADNI_nii_gz_bias_corrected/I387580_036_S_4894.stripped.N4.nii.gz
+DLBS0028592_session_1,F,40.97,dlbs_storage/DLBS_bias_corrected/0028592_session_1.stripped.N4.nii.gz
+ADNI293375,F,56.3,adni_storage/ADNI_nii_gz_bias_corrected/I293375_128_S_4607.stripped.N4.nii.gz
+fcon1000Beijing_sub92799,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub92799.stripped.N4.nii.gz
+DLBS0028354_session_1,M,75.42,dlbs_storage/DLBS_bias_corrected/0028354_session_1.stripped.N4.nii.gz
+sub-CC410287,F,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410287.stripped.N4.nii.gz
+fcon1000NewYork_sub44395,M,21.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub44395.stripped.N4.nii.gz
+ADNI62031,F,56.8,adni_storage/ADNI_nii_gz_bias_corrected/I62031_016_S_1121.stripped.N4.nii.gz
+fcon1000Beijing_sub31058,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub31058.stripped.N4.nii.gz
+ADNI1448643,F,62.9,adni_storage/ADNI_nii_gz_bias_corrected/I1448643_003_S_6954.stripped.N4.nii.gz
+sub-CC320461,M,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320461.stripped.N4.nii.gz
+ADNI1595258,F,53.5,adni_storage/ADNI_nii_gz_bias_corrected/I1595258_037_S_7035.stripped.N4.nii.gz
+fcon1000Beijing_sub29785,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub29785.stripped.N4.nii.gz
+fcon1000Cambridge_sub76631,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub76631.stripped.N4.nii.gz
+ADNI115775,F,59.2,adni_storage/ADNI_nii_gz_bias_corrected/I115775_033_S_0733.stripped.N4.nii.gz
+ADNI1143236,F,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I1143236_023_S_6702.stripped.N4.nii.gz
+DLBS0028459_session_1,M,66.13,dlbs_storage/DLBS_bias_corrected/0028459_session_1.stripped.N4.nii.gz
+ADNI415594,M,57.2,adni_storage/ADNI_nii_gz_bias_corrected/I415594_009_S_5252.stripped.N4.nii.gz
+sub-CC320379,M,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320379.stripped.N4.nii.gz
+fcon1000Beijing_sub08992,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub08992.stripped.N4.nii.gz
+fcon1000Beijing_sub50985,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub50985.stripped.N4.nii.gz
+sub-CC610671,F,70.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610671.stripped.N4.nii.gz
+fcon1000Beijing_sub89742,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub89742.stripped.N4.nii.gz
+DLBS0028397_session_1,F,35.33,dlbs_storage/DLBS_bias_corrected/0028397_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub82352,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub82352.stripped.N4.nii.gz
+fcon1000Beijing_sub64923,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub64923.stripped.N4.nii.gz
+sub-CC321291,F,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321291.stripped.N4.nii.gz
+fcon1000Cambridge_sub69315,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub69315.stripped.N4.nii.gz
+fcon1000Cambridge_sub77598,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub77598.stripped.N4.nii.gz
+sub-CC121685,M,20.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121685.stripped.N4.nii.gz
+sub-CC221244,F,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221244.stripped.N4.nii.gz
+DLBS0028538_session_1,F,76.39,dlbs_storage/DLBS_bias_corrected/0028538_session_1.stripped.N4.nii.gz
+fcon1000Leiden_sub28473,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub28473.stripped.N4.nii.gz
+sub-CC512003,M,62.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC512003.stripped.N4.nii.gz
+sub-CC510609,M,59.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510609.stripped.N4.nii.gz
+ADNI346264,F,61.0,adni_storage/ADNI_nii_gz_bias_corrected/I346264_036_S_4894.stripped.N4.nii.gz
+sub-CC520607,F,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520607.stripped.N4.nii.gz
+fcon1000Beijing_sub51586,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub51586.stripped.N4.nii.gz
+fcon1000Cambridge_sub78552,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub78552.stripped.N4.nii.gz
+fcon1000NewYork_sub59796,M,34.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub59796.stripped.N4.nii.gz
+fcon1000Baltimore_sub29158,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub29158.stripped.N4.nii.gz
+DLBS0028379_session_1,F,73.36,dlbs_storage/DLBS_bias_corrected/0028379_session_1.stripped.N4.nii.gz
+sub-CC320321,F,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320321.stripped.N4.nii.gz
+ADNI395985,F,59.3,adni_storage/ADNI_nii_gz_bias_corrected/I395985_130_S_4982.stripped.N4.nii.gz
+ADNI44672,M,60.1,adni_storage/ADNI_nii_gz_bias_corrected/I44672_020_S_1288.stripped.N4.nii.gz
+fcon1000Cambridge_sub19717,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub19717.stripped.N4.nii.gz
+sub-CC620164,F,72.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620164.stripped.N4.nii.gz
+DLBS0028412_session_1,M,20.89,dlbs_storage/DLBS_bias_corrected/0028412_session_1.stripped.N4.nii.gz
+ADNI285080,M,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I285080_073_S_4552.stripped.N4.nii.gz
+fcon1000Cleveland_sub18011,M,55.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub18011.stripped.N4.nii.gz
+sub-CC620152,M,73.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620152.stripped.N4.nii.gz
+ADNI1525978,F,64.1,adni_storage/ADNI_nii_gz_bias_corrected/I1525978_131_S_6805.stripped.N4.nii.gz
+fcon1000Beijing_sub82426,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub82426.stripped.N4.nii.gz
+DLBS0028438_session_1,M,85.99,dlbs_storage/DLBS_bias_corrected/0028438_session_1.stripped.N4.nii.gz
+ADNI460671,M,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I460671_073_S_4795.stripped.N4.nii.gz
+fcon1000Cambridge_sub20543,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub20543.stripped.N4.nii.gz
+DLBS0028596_session_1,M,36.92,dlbs_storage/DLBS_bias_corrected/0028596_session_1.stripped.N4.nii.gz
+sub-CC420464,F,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420464.stripped.N4.nii.gz
+sub-CC410086,M,57.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410086.stripped.N4.nii.gz
+sub-CC610568,M,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610568.stripped.N4.nii.gz
+DLBS0028622_session_1,F,24.6,dlbs_storage/DLBS_bias_corrected/0028622_session_1.stripped.N4.nii.gz
+ADNI1058589,M,63.4,adni_storage/ADNI_nii_gz_bias_corrected/I1058589_032_S_6602.stripped.N4.nii.gz
+sub-CC420204,F,53.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420204.stripped.N4.nii.gz
+DLBS0028537_session_1,F,62.2,dlbs_storage/DLBS_bias_corrected/0028537_session_1.stripped.N4.nii.gz
+ADNI871844,F,57.7,adni_storage/ADNI_nii_gz_bias_corrected/I871844_141_S_6041.stripped.N4.nii.gz
+ADNI324990,F,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I324990_023_S_4448.stripped.N4.nii.gz
+sub-CC220232,F,36.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220232.stripped.N4.nii.gz
+fcon1000Beijing_sub20127,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub20127.stripped.N4.nii.gz
+fcon1000Bangor_sub48632,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub48632.stripped.N4.nii.gz
+sub-CC720119,M,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720119.stripped.N4.nii.gz
+sub-CC420167,F,51.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420167.stripped.N4.nii.gz
+sub-CC620073,M,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620073.stripped.N4.nii.gz
+fcon1000Cambridge_sub92440,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub92440.stripped.N4.nii.gz
+ADNI1478765,F,56.9,adni_storage/ADNI_nii_gz_bias_corrected/I1478765_037_S_6977.stripped.N4.nii.gz
+sub-CC620106,F,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620106.stripped.N4.nii.gz
+sub-CC710982,M,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710982.stripped.N4.nii.gz
+ADNI1032325,F,56.5,adni_storage/ADNI_nii_gz_bias_corrected/I1032325_070_S_6548.stripped.N4.nii.gz
+DLBS0028453_session_1,F,73.26,dlbs_storage/DLBS_bias_corrected/0028453_session_1.stripped.N4.nii.gz
+DLBS0028473_session_1,F,56.11,dlbs_storage/DLBS_bias_corrected/0028473_session_1.stripped.N4.nii.gz
+fcon1000Newark_sub62985,F,25.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub62985.stripped.N4.nii.gz
+fcon1000AnnArbor_sub45569,F,63.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub45569.stripped.N4.nii.gz
+sub-CC510433,F,61.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510433.stripped.N4.nii.gz
+DLBS0028561_session_1,F,49.51,dlbs_storage/DLBS_bias_corrected/0028561_session_1.stripped.N4.nii.gz
+ADNI1331166,F,57.9,adni_storage/ADNI_nii_gz_bias_corrected/I1331166_116_S_6550.stripped.N4.nii.gz
+sub-CC610392,F,75.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610392.stripped.N4.nii.gz
+DLBS0028359_session_1,F,71.2,dlbs_storage/DLBS_bias_corrected/0028359_session_1.stripped.N4.nii.gz
+DLBS0028404_session_1,F,38.45,dlbs_storage/DLBS_bias_corrected/0028404_session_1.stripped.N4.nii.gz
+ADNI1072842,F,59.9,adni_storage/ADNI_nii_gz_bias_corrected/I1072842_014_S_6076.stripped.N4.nii.gz
+sub-CC310224,F,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310224.stripped.N4.nii.gz
+sub-CC220519,F,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220519.stripped.N4.nii.gz
+sub-CC310160,M,41.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310160.stripped.N4.nii.gz
+sub-CC620567,M,74.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620567.stripped.N4.nii.gz
+DLBS0028638_session_1,F,70.98,dlbs_storage/DLBS_bias_corrected/0028638_session_1.stripped.N4.nii.gz
+fcon1000NewYork_sub31554,F,24.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub31554.stripped.N4.nii.gz
+sub-CC520122,M,67.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520122.stripped.N4.nii.gz
+sub-CC320022,M,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320022.stripped.N4.nii.gz
+DLBS0028391_session_1,M,36.03,dlbs_storage/DLBS_bias_corrected/0028391_session_1.stripped.N4.nii.gz
+DLBS0028506_session_1,F,86.38,dlbs_storage/DLBS_bias_corrected/0028506_session_1.stripped.N4.nii.gz
+fcon1000Munchen_sub26223,M,65.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub26223.stripped.N4.nii.gz
+fcon1000Atlanta_sub15817,M,31.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub15817.stripped.N4.nii.gz
+DLBS0028465_session_1,F,21.67,dlbs_storage/DLBS_bias_corrected/0028465_session_1.stripped.N4.nii.gz
+fcon1000NewYork_sub56734,F,50.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub56734.stripped.N4.nii.gz
+ADNI1191310,M,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I1191310_130_S_6329.stripped.N4.nii.gz
+DLBS0028554_session_1,F,48.67,dlbs_storage/DLBS_bias_corrected/0028554_session_1.stripped.N4.nii.gz
+sub-CC510415,F,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510415.stripped.N4.nii.gz
+fcon1000Newark_sub13411,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub13411.stripped.N4.nii.gz
+fcon1000Cambridge_sub07902,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub07902.stripped.N4.nii.gz
+sub-CC220526,F,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220526.stripped.N4.nii.gz
+sub-CC710037,F,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710037.stripped.N4.nii.gz
+ADNI1072841,F,59.9,adni_storage/ADNI_nii_gz_bias_corrected/I1072841_014_S_6076.stripped.N4.nii.gz
+fcon1000Cambridge_sub60578,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub60578.stripped.N4.nii.gz
+fcon1000Cambridge_sub08723,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub08723.stripped.N4.nii.gz
+DLBS0028535_session_1,F,80.11,dlbs_storage/DLBS_bias_corrected/0028535_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub06272,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub06272.stripped.N4.nii.gz
+ADNI1575205,F,62.6,adni_storage/ADNI_nii_gz_bias_corrected/I1575205_006_S_6651.stripped.N4.nii.gz
+ADNI10666923,F,64.6,adni_storage/ADNI_nii_gz_bias_corrected/I10666923_100_S_10091.stripped.N4.nii.gz
+ADNI323094,F,60.8,adni_storage/ADNI_nii_gz_bias_corrected/I323094_036_S_4894.stripped.N4.nii.gz
+fcon1000Cambridge_sub13902,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub13902.stripped.N4.nii.gz
+fcon1000Cambridge_sub49259,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub49259.stripped.N4.nii.gz
+fcon1000Cambridge_sub65373,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub65373.stripped.N4.nii.gz
+fcon1000Beijing_sub38602,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub38602.stripped.N4.nii.gz
+ADNI400951,M,62.8,adni_storage/ADNI_nii_gz_bias_corrected/I400951_073_S_4795.stripped.N4.nii.gz
+fcon1000Berlin_sub27711,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub27711.stripped.N4.nii.gz
+fcon1000NewYork_sub33581,M,49.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub33581.stripped.N4.nii.gz
+sub-CC120727,F,23.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120727.stripped.N4.nii.gz
+sub-CC620592,F,74.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620592.stripped.N4.nii.gz
+fcon1000Bangor_sub01903,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub01903.stripped.N4.nii.gz
+fcon1000Cambridge_sub53193,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub53193.stripped.N4.nii.gz
+sub-CC220335,M,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220335.stripped.N4.nii.gz
+sub-CC610405,F,72.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610405.stripped.N4.nii.gz
+sub-CC221595,M,33.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221595.stripped.N4.nii.gz
+DLBS0028488_session_1,F,41.33,dlbs_storage/DLBS_bias_corrected/0028488_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub44573,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub44573.stripped.N4.nii.gz
+fcon1000Munchen_sub31272,M,74.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub31272.stripped.N4.nii.gz
+DLBS0028335_session_1,M,71.34,dlbs_storage/DLBS_bias_corrected/0028335_session_1.stripped.N4.nii.gz
+fcon1000Atlanta_sub00354,F,28.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub00354.stripped.N4.nii.gz
+DLBS0028617_session_1,F,50.6,dlbs_storage/DLBS_bias_corrected/0028617_session_1.stripped.N4.nii.gz
+sub-CC520390,F,65.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520390.stripped.N4.nii.gz
+sub-CC620885,M,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620885.stripped.N4.nii.gz
+fcon1000Cambridge_sub31522,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub31522.stripped.N4.nii.gz
+ADNI1185293,F,60.0,adni_storage/ADNI_nii_gz_bias_corrected/I1185293_024_S_6033.stripped.N4.nii.gz
+sub-CC210657,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210657.stripped.N4.nii.gz
+sub-CC321025,M,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321025.stripped.N4.nii.gz
+ADNI381380,M,56.6,adni_storage/ADNI_nii_gz_bias_corrected/I381380_009_S_5252.stripped.N4.nii.gz
+fcon1000Beijing_sub80569,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub80569.stripped.N4.nii.gz
+sub-CC620619,M,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620619.stripped.N4.nii.gz
+sub-CC222258,F,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222258.stripped.N4.nii.gz
+DLBS0028613_session_1,M,74.4,dlbs_storage/DLBS_bias_corrected/0028613_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub66506,F,78.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub66506.stripped.N4.nii.gz
+DLBS0028549_session_1,F,21.45,dlbs_storage/DLBS_bias_corrected/0028549_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub95187,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub95187.stripped.N4.nii.gz
+ADNI1485046,F,58.9,adni_storage/ADNI_nii_gz_bias_corrected/I1485046_116_S_6550.stripped.N4.nii.gz
+sub-CC320574,M,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320574.stripped.N4.nii.gz
+sub-CC620444,F,77.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620444.stripped.N4.nii.gz
+sub-CC510243,M,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510243.stripped.N4.nii.gz
+sub-CC610508,F,77.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610508.stripped.N4.nii.gz
+fcon1000Milwaukee_sub18955,F,53.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub18955.stripped.N4.nii.gz
+sub-CC321557,F,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321557.stripped.N4.nii.gz
+sub-CC510629,M,59.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510629.stripped.N4.nii.gz
+DLBS0028604_session_1,F,22.6,dlbs_storage/DLBS_bias_corrected/0028604_session_1.stripped.N4.nii.gz
+ADNI1227908,F,61.1,adni_storage/ADNI_nii_gz_bias_corrected/I1227908_041_S_6801.stripped.N4.nii.gz
+sub-CC721729,M,81.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721729.stripped.N4.nii.gz
+sub-CC320325,F,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320325.stripped.N4.nii.gz
+sub-CC221511,F,36.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221511.stripped.N4.nii.gz
+sub-CC610076,F,77.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610076.stripped.N4.nii.gz
+sub-CC520980,M,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520980.stripped.N4.nii.gz
+DLBS0028478_session_1,F,70.35,dlbs_storage/DLBS_bias_corrected/0028478_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub46058,M,25.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub46058.stripped.N4.nii.gz
+ADNI228875,F,56.0,adni_storage/ADNI_nii_gz_bias_corrected/I228875_031_S_4024.stripped.N4.nii.gz
+DLBS0028601_session_1,M,71.54,dlbs_storage/DLBS_bias_corrected/0028601_session_1.stripped.N4.nii.gz
+fcon1000Cleveland_sub26557,M,51.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub26557.stripped.N4.nii.gz
+sub-CC410032,M,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410032.stripped.N4.nii.gz
+ADNI1461224,M,61.6,adni_storage/ADNI_nii_gz_bias_corrected/I1461224_070_S_6966.stripped.N4.nii.gz
+sub-CC320336,F,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320336.stripped.N4.nii.gz
+sub-CC620354,F,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620354.stripped.N4.nii.gz
+fcon1000Leiden_sub92061,M,28.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub92061.stripped.N4.nii.gz
+DLBS0028344_session_1,F,69.81,dlbs_storage/DLBS_bias_corrected/0028344_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub95644,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub95644.stripped.N4.nii.gz
+ADNI1215082,F,56.9,adni_storage/ADNI_nii_gz_bias_corrected/I1215082_116_S_6550.stripped.N4.nii.gz
+sub-CC110187,F,25.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110187.stripped.N4.nii.gz
+fcon1000Baltimore_sub96234,F,30.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub96234.stripped.N4.nii.gz
+fcon1000Leiden_sub56299,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub56299.stripped.N4.nii.gz
+ADNI400958,M,62.8,adni_storage/ADNI_nii_gz_bias_corrected/I400958_073_S_4795.stripped.N4.nii.gz
+fcon1000Beijing_sub08001,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub08001.stripped.N4.nii.gz
+fcon1000Leiden_sub93194,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub93194.stripped.N4.nii.gz
+DLBS0028508_session_1,F,28.67,dlbs_storage/DLBS_bias_corrected/0028508_session_1.stripped.N4.nii.gz
+fcon1000Leiden_sub01787,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub01787.stripped.N4.nii.gz
+sub-CC320776,M,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320776.stripped.N4.nii.gz
+sub-CC121111,M,18.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121111.stripped.N4.nii.gz
+fcon1000Cambridge_sub69397,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub69397.stripped.N4.nii.gz
+fcon1000AnnArbor_sub99692,F,66.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub99692.stripped.N4.nii.gz
+fcon1000Newark_sub59799,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub59799.stripped.N4.nii.gz
+sub-CC321368,M,45.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321368.stripped.N4.nii.gz
+ADNI381382,M,56.6,adni_storage/ADNI_nii_gz_bias_corrected/I381382_009_S_5252.stripped.N4.nii.gz
+ADNI90757,F,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I90757_051_S_1072.stripped.N4.nii.gz
+fcon1000AnnArbor_sub57025,F,68.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub57025.stripped.N4.nii.gz
+sub-CC320575,F,39.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320575.stripped.N4.nii.gz
+ADNI323095,F,60.8,adni_storage/ADNI_nii_gz_bias_corrected/I323095_036_S_4894.stripped.N4.nii.gz
+sub-CC710088,F,83.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710088.stripped.N4.nii.gz
+sub-CC710518,F,83.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710518.stripped.N4.nii.gz
+DLBS0028587_session_1,F,72.59,dlbs_storage/DLBS_bias_corrected/0028587_session_1.stripped.N4.nii.gz
+DLBS0028331_session_1,F,77.51,dlbs_storage/DLBS_bias_corrected/0028331_session_1.stripped.N4.nii.gz
+fcon1000Newark_sub58526,F,25.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub58526.stripped.N4.nii.gz
+ADNI362186,F,58.4,adni_storage/ADNI_nii_gz_bias_corrected/I362186_041_S_4874.stripped.N4.nii.gz
+sub-CC510258,M,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510258.stripped.N4.nii.gz
+sub-CC520673,M,67.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520673.stripped.N4.nii.gz
+DLBS0028484_session_1,M,70.08,dlbs_storage/DLBS_bias_corrected/0028484_session_1.stripped.N4.nii.gz
+fcon1000Munchen_sub26670,M,64.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub26670.stripped.N4.nii.gz
+fcon1000Cambridge_sub25044,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub25044.stripped.N4.nii.gz
+fcon1000NewYork_sub17078,M,34.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub17078.stripped.N4.nii.gz
+sub-CC410015,F,51.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410015.stripped.N4.nii.gz
+sub-CC410084,F,57.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410084.stripped.N4.nii.gz
+sub-CC520136,F,65.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520136.stripped.N4.nii.gz
+fcon1000Berlin_sub77281,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub77281.stripped.N4.nii.gz
+fcon1000Milwaukee_sub53971,F,47.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub53971.stripped.N4.nii.gz
+sub-CC621128,M,75.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC621128.stripped.N4.nii.gz
+sub-CC121428,F,26.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121428.stripped.N4.nii.gz
+ADNI10666919,F,59.7,adni_storage/ADNI_nii_gz_bias_corrected/I10666919_301_S_10139.stripped.N4.nii.gz
+sub-CC621199,M,72.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC621199.stripped.N4.nii.gz
+fcon1000Beijing_sub93689,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub93689.stripped.N4.nii.gz
+ADNI1080321,F,62.9,adni_storage/ADNI_nii_gz_bias_corrected/I1080321_114_S_6597.stripped.N4.nii.gz
+DLBS0028434_session_1,F,70.1,dlbs_storage/DLBS_bias_corrected/0028434_session_1.stripped.N4.nii.gz
+DLBS0028576_session_1,F,32.92,dlbs_storage/DLBS_bias_corrected/0028576_session_1.stripped.N4.nii.gz
+DLBS0028342_session_1,F,31.98,dlbs_storage/DLBS_bias_corrected/0028342_session_1.stripped.N4.nii.gz
+ADNI460666,M,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I460666_073_S_4795.stripped.N4.nii.gz
+sub-CC420222,M,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420222.stripped.N4.nii.gz
+sub-CC720685,M,81.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720685.stripped.N4.nii.gz
+sub-CC221209,F,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221209.stripped.N4.nii.gz
+fcon1000Milwaukee_sub09931,M,48.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub09931.stripped.N4.nii.gz
+fcon1000Cambridge_sub14278,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub14278.stripped.N4.nii.gz
+fcon1000Cambridge_sub29425,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub29425.stripped.N4.nii.gz
+ADNI1195772,F,63.7,adni_storage/ADNI_nii_gz_bias_corrected/I1195772_019_S_6483.stripped.N4.nii.gz
+fcon1000Cambridge_sub55874,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub55874.stripped.N4.nii.gz
+ADNI274149,M,63.0,adni_storage/ADNI_nii_gz_bias_corrected/I274149_002_S_1155.stripped.N4.nii.gz
+ADNI10302489,M,58.5,adni_storage/ADNI_nii_gz_bias_corrected/I10302489_011_S_7112.stripped.N4.nii.gz
+fcon1000NewYork_sub23844,F,20.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub23844.stripped.N4.nii.gz
+DLBS0028515_session_1,F,50.9,dlbs_storage/DLBS_bias_corrected/0028515_session_1.stripped.N4.nii.gz
+ADNI1253903,M,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I1253903_032_S_6602.stripped.N4.nii.gz
+fcon1000Beijing_sub06880,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub06880.stripped.N4.nii.gz
+sub-CC120313,M,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120313.stripped.N4.nii.gz
+ADNI362185,F,58.4,adni_storage/ADNI_nii_gz_bias_corrected/I362185_041_S_4874.stripped.N4.nii.gz
+ADNI10874412,M,63.0,adni_storage/ADNI_nii_gz_bias_corrected/I10874412_019_S_10145.stripped.N4.nii.gz
+ADNI395983,F,59.3,adni_storage/ADNI_nii_gz_bias_corrected/I395983_130_S_4982.stripped.N4.nii.gz
+fcon1000Baltimore_sub91622,M,33.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub91622.stripped.N4.nii.gz
+DLBS0028414_session_1,F,68.45,dlbs_storage/DLBS_bias_corrected/0028414_session_1.stripped.N4.nii.gz
+ADNI114056,F,57.9,adni_storage/ADNI_nii_gz_bias_corrected/I114056_016_S_1121.stripped.N4.nii.gz
+sub-CC510395,M,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510395.stripped.N4.nii.gz
+fcon1000Cambridge_sub89107,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub89107.stripped.N4.nii.gz
+fcon1000Cambridge_sub60797,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub60797.stripped.N4.nii.gz
+fcon1000Newark_sub55760,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub55760.stripped.N4.nii.gz
+DLBS0028446_session_1,M,50.89,dlbs_storage/DLBS_bias_corrected/0028446_session_1.stripped.N4.nii.gz
+fcon1000Leiden_sub87320,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub87320.stripped.N4.nii.gz
+DLBS0028608_session_1,F,38.13,dlbs_storage/DLBS_bias_corrected/0028608_session_1.stripped.N4.nii.gz
+sub-CC321331,M,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321331.stripped.N4.nii.gz
+fcon1000Beijing_sub34895,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub34895.stripped.N4.nii.gz
+sub-CC220697,M,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220697.stripped.N4.nii.gz
+ADNI1454373,F,55.8,adni_storage/ADNI_nii_gz_bias_corrected/I1454373_137_S_6906.stripped.N4.nii.gz
+ADNI844181,F,60.4,adni_storage/ADNI_nii_gz_bias_corrected/I844181_022_S_6013.stripped.N4.nii.gz
+sub-CC621184,M,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC621184.stripped.N4.nii.gz
+sub-CC510473,M,65.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510473.stripped.N4.nii.gz
+DLBS0028370_session_1,F,22.88,dlbs_storage/DLBS_bias_corrected/0028370_session_1.stripped.N4.nii.gz
+DLBS0028567_session_1,F,85.6,dlbs_storage/DLBS_bias_corrected/0028567_session_1.stripped.N4.nii.gz
+fcon1000NewYork_sub01912,M,31.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub01912.stripped.N4.nii.gz
+ADNI1469814,M,57.0,adni_storage/ADNI_nii_gz_bias_corrected/I1469814_073_S_6925.stripped.N4.nii.gz
+sub-CC410222,F,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410222.stripped.N4.nii.gz
+DLBS0028445_session_1,M,55.37,dlbs_storage/DLBS_bias_corrected/0028445_session_1.stripped.N4.nii.gz
+DLBS0028451_session_1,F,43.61,dlbs_storage/DLBS_bias_corrected/0028451_session_1.stripped.N4.nii.gz
+DLBS0028503_session_1,F,88.32,dlbs_storage/DLBS_bias_corrected/0028503_session_1.stripped.N4.nii.gz
+sub-CC221220,F,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221220.stripped.N4.nii.gz
+sub-CC310203,F,39.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310203.stripped.N4.nii.gz
+ADNI73490,F,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I73490_067_S_1253.stripped.N4.nii.gz
+fcon1000Beijing_sub04050,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub04050.stripped.N4.nii.gz
+ADNI168101,F,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I168101_051_S_1072.stripped.N4.nii.gz
+sub-CC420094,M,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420094.stripped.N4.nii.gz
+fcon1000Cambridge_sub15172,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub15172.stripped.N4.nii.gz
+fcon1000Cambridge_sub14183,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub14183.stripped.N4.nii.gz
+sub-CC220506,F,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220506.stripped.N4.nii.gz
+sub-CC722542,M,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC722542.stripped.N4.nii.gz
+DLBS0028401_session_1,F,54.05,dlbs_storage/DLBS_bias_corrected/0028401_session_1.stripped.N4.nii.gz
+sub-CC721891,M,83.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721891.stripped.N4.nii.gz
+sub-CC510161,M,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510161.stripped.N4.nii.gz
+sub-CC621011,M,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC621011.stripped.N4.nii.gz
+sub-CC520209,F,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520209.stripped.N4.nii.gz
+DLBS0028625_session_1,F,60.37,dlbs_storage/DLBS_bias_corrected/0028625_session_1.stripped.N4.nii.gz
+ADNI60863,F,57.3,adni_storage/ADNI_nii_gz_bias_corrected/I60863_133_S_0913.stripped.N4.nii.gz
+ADNI1559412,F,60.2,adni_storage/ADNI_nii_gz_bias_corrected/I1559412_082_S_6629.stripped.N4.nii.gz
+sub-CC220920,F,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220920.stripped.N4.nii.gz
+fcon1000Cambridge_sub83409,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub83409.stripped.N4.nii.gz
+ADNI1406523,F,55.5,adni_storage/ADNI_nii_gz_bias_corrected/I1406523_305_S_6899.stripped.N4.nii.gz
+fcon1000Berlin_sub67166,F,32.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub67166.stripped.N4.nii.gz
+sub-CC221336,M,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221336.stripped.N4.nii.gz
+fcon1000Beijing_sub07717,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub07717.stripped.N4.nii.gz
+fcon1000AnnArbor_sub05580,F,77.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub05580.stripped.N4.nii.gz
+DLBS0028386_session_1,F,33.13,dlbs_storage/DLBS_bias_corrected/0028386_session_1.stripped.N4.nii.gz
+ADNI10386604,F,57.2,adni_storage/ADNI_nii_gz_bias_corrected/I10386604_035_S_10067.stripped.N4.nii.gz
+fcon1000Beijing_sub68012,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub68012.stripped.N4.nii.gz
+ADNI1194377,F,55.1,adni_storage/ADNI_nii_gz_bias_corrected/I1194377_016_S_6773.stripped.N4.nii.gz
+sub-CC410113,M,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410113.stripped.N4.nii.gz
+sub-CC120264,M,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120264.stripped.N4.nii.gz
+ADNI347410,F,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I347410_130_S_4997.stripped.N4.nii.gz
+ADNI1254370,F,60.9,adni_storage/ADNI_nii_gz_bias_corrected/I1254370_014_S_6076.stripped.N4.nii.gz
+ADNI423069,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I423069_128_S_4607.stripped.N4.nii.gz
+DLBS0028476_session_1,F,65.91,dlbs_storage/DLBS_bias_corrected/0028476_session_1.stripped.N4.nii.gz
+sub-CC510474,M,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510474.stripped.N4.nii.gz
+fcon1000Milwaukee_sub56333,M,53.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub56333.stripped.N4.nii.gz
+sub-CC221373,M,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221373.stripped.N4.nii.gz
+DLBS0028362_session_1,F,60.87,dlbs_storage/DLBS_bias_corrected/0028362_session_1.stripped.N4.nii.gz
+sub-CC510015,F,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510015.stripped.N4.nii.gz
+sub-CC510438,M,62.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510438.stripped.N4.nii.gz
+ADNI974338,M,62.1,adni_storage/ADNI_nii_gz_bias_corrected/I974338_003_S_6257.stripped.N4.nii.gz
+sub-CC320636,M,45.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320636.stripped.N4.nii.gz
+sub-CC520395,F,61.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520395.stripped.N4.nii.gz
+fcon1000Berlin_sub33248,F,28.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub33248.stripped.N4.nii.gz
+sub-CC210088,F,36.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210088.stripped.N4.nii.gz
+DLBS0028472_session_1,M,42.45,dlbs_storage/DLBS_bias_corrected/0028472_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub39737,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub39737.stripped.N4.nii.gz
+ADNI344059,F,58.0,adni_storage/ADNI_nii_gz_bias_corrected/I344059_041_S_4874.stripped.N4.nii.gz
+fcon1000Milwaukee_sub45852,F,63.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub45852.stripped.N4.nii.gz
+fcon1000Beijing_sub22715,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub22715.stripped.N4.nii.gz
+fcon1000Beijing_sub17093,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub17093.stripped.N4.nii.gz
+fcon1000NewYork_sub28795,M,10.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub28795.stripped.N4.nii.gz
+DLBS0028544_session_1,M,81.75,dlbs_storage/DLBS_bias_corrected/0028544_session_1.stripped.N4.nii.gz
+sub-CC520134,F,67.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520134.stripped.N4.nii.gz
+ADNI10937009,M,59.1,adni_storage/ADNI_nii_gz_bias_corrected/I10937009_052_S_10188.stripped.N4.nii.gz
+ADNI132477,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I132477_016_S_1121.stripped.N4.nii.gz
+sub-CC721519,F,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721519.stripped.N4.nii.gz
+ADNI10454377,M,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I10454377_022_S_10055.stripped.N4.nii.gz
+sub-CC222185,F,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222185.stripped.N4.nii.gz
+ADNI1446679,F,59.8,adni_storage/ADNI_nii_gz_bias_corrected/I1446679_070_S_6942.stripped.N4.nii.gz
+fcon1000AnnArbor_sub85257,M,26.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub85257.stripped.N4.nii.gz
+fcon1000NewYork_sub29216,M,29.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub29216.stripped.N4.nii.gz
+DLBS0028479_session_1,M,40.93,dlbs_storage/DLBS_bias_corrected/0028479_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub35776,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub35776.stripped.N4.nii.gz
+sub-CC210023,M,31.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210023.stripped.N4.nii.gz
+ADNI267809,F,64.1,adni_storage/ADNI_nii_gz_bias_corrected/I267809_037_S_4028.stripped.N4.nii.gz
+fcon1000NewYork_sub53461,M,27.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub53461.stripped.N4.nii.gz
+fcon1000Milwaukee_sub55176,F,63.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub55176.stripped.N4.nii.gz
+DLBS0028420_session_1,F,81.2,dlbs_storage/DLBS_bias_corrected/0028420_session_1.stripped.N4.nii.gz
+sub-CC510237,M,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510237.stripped.N4.nii.gz
+fcon1000Cambridge_sub49998,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub49998.stripped.N4.nii.gz
+fcon1000Cambridge_sub13187,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub13187.stripped.N4.nii.gz
+sub-CC222496,M,38.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222496.stripped.N4.nii.gz
+fcon1000Cambridge_sub04491,F,25.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub04491.stripped.N4.nii.gz
+sub-CC120376,F,18.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120376.stripped.N4.nii.gz
+ADNI114263,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I114263_133_S_0913.stripped.N4.nii.gz
+DLBS0028602_session_1,F,29.19,dlbs_storage/DLBS_bias_corrected/0028602_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub15846,F,70.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub15846.stripped.N4.nii.gz
+fcon1000Baltimore_sub76160,M,34.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub76160.stripped.N4.nii.gz
+fcon1000Cambridge_sub99462,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub99462.stripped.N4.nii.gz
+fcon1000Beijing_sub66889,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub66889.stripped.N4.nii.gz
+ADNI87682,F,57.8,adni_storage/ADNI_nii_gz_bias_corrected/I87682_133_S_0913.stripped.N4.nii.gz
+ADNI1254369,F,60.9,adni_storage/ADNI_nii_gz_bias_corrected/I1254369_014_S_6076.stripped.N4.nii.gz
+ADNI367180,M,61.8,adni_storage/ADNI_nii_gz_bias_corrected/I367180_135_S_4954.stripped.N4.nii.gz
+fcon1000Berlin_sub54976,M,37.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub54976.stripped.N4.nii.gz
+fcon1000Beijing_sub87776,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub87776.stripped.N4.nii.gz
+fcon1000Milwaukee_sub59359,F,51.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub59359.stripped.N4.nii.gz
+ADNI1079629,F,57.0,adni_storage/ADNI_nii_gz_bias_corrected/I1079629_082_S_6629.stripped.N4.nii.gz
+sub-CC221352,F,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221352.stripped.N4.nii.gz
+sub-CC210124,F,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210124.stripped.N4.nii.gz
+fcon1000NewYork_sub19579,M,46.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub19579.stripped.N4.nii.gz
+sub-CC320568,F,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320568.stripped.N4.nii.gz
+sub-CC520503,M,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520503.stripped.N4.nii.gz
+fcon1000Baltimore_sub77572,F,27.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub77572.stripped.N4.nii.gz
+DLBS0028375_session_1,F,59.07,dlbs_storage/DLBS_bias_corrected/0028375_session_1.stripped.N4.nii.gz
+fcon1000Milwaukee_sub30157,M,54.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub30157.stripped.N4.nii.gz
+fcon1000AnnArbor_sub78151,M,23.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub78151.stripped.N4.nii.gz
+DLBS0028586_session_1,M,79.85,dlbs_storage/DLBS_bias_corrected/0028586_session_1.stripped.N4.nii.gz
+DLBS0028447_session_1,M,51.78,dlbs_storage/DLBS_bias_corrected/0028447_session_1.stripped.N4.nii.gz
+sub-CC221954,F,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221954.stripped.N4.nii.gz
+sub-CC320893,F,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320893.stripped.N4.nii.gz
+fcon1000Beijing_sub07144,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub07144.stripped.N4.nii.gz
+DLBS0028482_session_1,M,63.47,dlbs_storage/DLBS_bias_corrected/0028482_session_1.stripped.N4.nii.gz
+DLBS0028349_session_1,M,64.17,dlbs_storage/DLBS_bias_corrected/0028349_session_1.stripped.N4.nii.gz
+fcon1000Munchen_sub81737,F,67.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub81737.stripped.N4.nii.gz
+ADNI341153,M,60.6,adni_storage/ADNI_nii_gz_bias_corrected/I341153_128_S_4599.stripped.N4.nii.gz
+sub-CC110098,M,23.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110098.stripped.N4.nii.gz
+sub-CC621118,M,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC621118.stripped.N4.nii.gz
+DLBS0028452_session_1,F,59.71,dlbs_storage/DLBS_bias_corrected/0028452_session_1.stripped.N4.nii.gz
+ADNI228845,F,63.5,adni_storage/ADNI_nii_gz_bias_corrected/I228845_037_S_4028.stripped.N4.nii.gz
+DLBS0028614_session_1,F,53.77,dlbs_storage/DLBS_bias_corrected/0028614_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub24670,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub24670.stripped.N4.nii.gz
+ADNI1181048,M,60.0,adni_storage/ADNI_nii_gz_bias_corrected/I1181048_116_S_6428.stripped.N4.nii.gz
+fcon1000Cambridge_sub59729,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub59729.stripped.N4.nii.gz
+DLBS0028425_session_1,M,23.02,dlbs_storage/DLBS_bias_corrected/0028425_session_1.stripped.N4.nii.gz
+sub-CC510342,F,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510342.stripped.N4.nii.gz
+fcon1000Milwaukee_sub58967,F,61.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub58967.stripped.N4.nii.gz
+sub-CC220107,M,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220107.stripped.N4.nii.gz
+sub-CC620499,M,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620499.stripped.N4.nii.gz
+fcon1000Cambridge_sub89435,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub89435.stripped.N4.nii.gz
+sub-CC420162,F,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420162.stripped.N4.nii.gz
+fcon1000Newark_sub78225,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub78225.stripped.N4.nii.gz
+fcon1000AnnArbor_sub38614,M,14.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub38614.stripped.N4.nii.gz
+fcon1000Leiden_sub52853,M,27.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub52853.stripped.N4.nii.gz
+DLBS0028353_session_1,M,60.01,dlbs_storage/DLBS_bias_corrected/0028353_session_1.stripped.N4.nii.gz
+ADNI10471094,F,58.8,adni_storage/ADNI_nii_gz_bias_corrected/I10471094_073_S_10113.stripped.N4.nii.gz
+sub-CC312058,M,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC312058.stripped.N4.nii.gz
+sub-CC710176,F,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710176.stripped.N4.nii.gz
+ADNI296147,M,63.9,adni_storage/ADNI_nii_gz_bias_corrected/I296147_116_S_4625.stripped.N4.nii.gz
+sub-CC520562,M,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520562.stripped.N4.nii.gz
+DLBS0028621_session_1,F,30.6,dlbs_storage/DLBS_bias_corrected/0028621_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub53269,M,21.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub53269.stripped.N4.nii.gz
+DLBS0028529_session_1,F,86.77,dlbs_storage/DLBS_bias_corrected/0028529_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub30272,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub30272.stripped.N4.nii.gz
+fcon1000Beijing_sub89088,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub89088.stripped.N4.nii.gz
+DLBS0028396_session_1,M,67.74,dlbs_storage/DLBS_bias_corrected/0028396_session_1.stripped.N4.nii.gz
+sub-CC620479,M,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620479.stripped.N4.nii.gz
+ADNI352951,M,64.0,adni_storage/ADNI_nii_gz_bias_corrected/I352951_002_S_1155.stripped.N4.nii.gz
+fcon1000Milwaukee_sub64463,M,52.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub64463.stripped.N4.nii.gz
+sub-CC620129,M,75.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620129.stripped.N4.nii.gz
+DLBS0028594_session_1,M,80.29,dlbs_storage/DLBS_bias_corrected/0028594_session_1.stripped.N4.nii.gz
+sub-CC220974,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220974.stripped.N4.nii.gz
+sub-CC321281,F,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321281.stripped.N4.nii.gz
+sub-CC310252,M,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310252.stripped.N4.nii.gz
+ADNI944422,M,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I944422_041_S_6136.stripped.N4.nii.gz
+fcon1000Cambridge_sub61753,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub61753.stripped.N4.nii.gz
+sub-CC310129,M,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310129.stripped.N4.nii.gz
+sub-CC320088,M,41.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320088.stripped.N4.nii.gz
+sub-CC621642,M,73.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC621642.stripped.N4.nii.gz
+fcon1000Cambridge_sub17772,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub17772.stripped.N4.nii.gz
+sub-CC321174,F,39.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321174.stripped.N4.nii.gz
+ADNI1486321,M,57.1,adni_storage/ADNI_nii_gz_bias_corrected/I1486321_941_S_6998.stripped.N4.nii.gz
+ADNI313534,M,64.1,adni_storage/ADNI_nii_gz_bias_corrected/I313534_116_S_4625.stripped.N4.nii.gz
+DLBS0028340_session_1,F,67.83,dlbs_storage/DLBS_bias_corrected/0028340_session_1.stripped.N4.nii.gz
+sub-CC321431,F,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321431.stripped.N4.nii.gz
+sub-CC711158,M,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC711158.stripped.N4.nii.gz
+sub-CC420412,M,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420412.stripped.N4.nii.gz
+sub-CC510115,F,62.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510115.stripped.N4.nii.gz
+sub-CC120208,F,24.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120208.stripped.N4.nii.gz
+fcon1000NewYork_sub41546,F,24.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub41546.stripped.N4.nii.gz
+sub-CC610372,M,70.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610372.stripped.N4.nii.gz
+DLBS0028350_session_1,F,49.82,dlbs_storage/DLBS_bias_corrected/0028350_session_1.stripped.N4.nii.gz
+sub-CC620785,M,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620785.stripped.N4.nii.gz
+sub-CC710382,M,83.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710382.stripped.N4.nii.gz
+DLBS0028352_session_1,F,29.84,dlbs_storage/DLBS_bias_corrected/0028352_session_1.stripped.N4.nii.gz
+ADNI296878,F,58.5,adni_storage/ADNI_nii_gz_bias_corrected/I296878_012_S_4642.stripped.N4.nii.gz
+fcon1000Munchen_sub96752,F,63.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub96752.stripped.N4.nii.gz
+ADNI1514331,F,52.8,adni_storage/ADNI_nii_gz_bias_corrected/I1514331_037_S_6992.stripped.N4.nii.gz
+ADNI368465,M,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I368465_073_S_4552.stripped.N4.nii.gz
+fcon1000Baltimore_sub90893,F,39.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub90893.stripped.N4.nii.gz
+fcon1000Cambridge_sub82435,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub82435.stripped.N4.nii.gz
+fcon1000Beijing_sub15441,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub15441.stripped.N4.nii.gz
+sub-CC621284,M,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC621284.stripped.N4.nii.gz
+sub-CC410354,M,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410354.stripped.N4.nii.gz
+sub-CC122405,M,27.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC122405.stripped.N4.nii.gz
+fcon1000Cambridge_sub50272,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub50272.stripped.N4.nii.gz
+sub-CC721707,F,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721707.stripped.N4.nii.gz
+fcon1000Bangor_sub04097,M,36.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub04097.stripped.N4.nii.gz
+ADNI1092240,M,57.4,adni_storage/ADNI_nii_gz_bias_corrected/I1092240_114_S_6039.stripped.N4.nii.gz
+DLBS0028631_session_1,F,29.92,dlbs_storage/DLBS_bias_corrected/0028631_session_1.stripped.N4.nii.gz
+fcon1000Milwaukee_sub87784,M,58.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub87784.stripped.N4.nii.gz
+ADNI365455,M,62.1,adni_storage/ADNI_nii_gz_bias_corrected/I365455_073_S_4795.stripped.N4.nii.gz
+sub-CC310414,M,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310414.stripped.N4.nii.gz
+fcon1000Cleveland_sub02480,F,56.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub02480.stripped.N4.nii.gz
+fcon1000Milwaukee_sub24237,F,52.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub24237.stripped.N4.nii.gz
+fcon1000NewYork_sub15213,M,16.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub15213.stripped.N4.nii.gz
+sub-CC110037,M,18.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110037.stripped.N4.nii.gz
+fcon1000Beijing_sub71693,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub71693.stripped.N4.nii.gz
+DLBS0028490_session_1,F,63.73,dlbs_storage/DLBS_bias_corrected/0028490_session_1.stripped.N4.nii.gz
+sub-CC321880,F,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321880.stripped.N4.nii.gz
+sub-CC321594,M,43.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321594.stripped.N4.nii.gz
+sub-CC221828,F,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221828.stripped.N4.nii.gz
+DLBS0028474_session_1,F,83.7,dlbs_storage/DLBS_bias_corrected/0028474_session_1.stripped.N4.nii.gz
+ADNI247842,F,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I247842_037_S_4028.stripped.N4.nii.gz
+fcon1000Cambridge_sub57221,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub57221.stripped.N4.nii.gz
+sub-CC120286,M,22.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120286.stripped.N4.nii.gz
+DLBS0028441_session_1,M,45.18,dlbs_storage/DLBS_bias_corrected/0028441_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub28698,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub28698.stripped.N4.nii.gz
+ADNI117521,F,61.7,adni_storage/ADNI_nii_gz_bias_corrected/I117521_051_S_1072.stripped.N4.nii.gz
+fcon1000Cambridge_sub47162,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub47162.stripped.N4.nii.gz
+fcon1000Beijing_sub54890,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub54890.stripped.N4.nii.gz
+sub-CC720304,F,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720304.stripped.N4.nii.gz
+fcon1000Cambridge_sub58682,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub58682.stripped.N4.nii.gz
+fcon1000Beijing_sub05267,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub05267.stripped.N4.nii.gz
+ADNI36695,F,56.4,adni_storage/ADNI_nii_gz_bias_corrected/I36695_016_S_1121.stripped.N4.nii.gz
+sub-CC520775,F,61.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520775.stripped.N4.nii.gz
+fcon1000Beijing_sub81062,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub81062.stripped.N4.nii.gz
+fcon1000Milwaukee_sub58677,F,46.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub58677.stripped.N4.nii.gz
+DLBS0028374_session_1,F,55.62,dlbs_storage/DLBS_bias_corrected/0028374_session_1.stripped.N4.nii.gz
+DLBS0028345_session_1,F,59.73,dlbs_storage/DLBS_bias_corrected/0028345_session_1.stripped.N4.nii.gz
+DLBS0028381_session_1,F,24.24,dlbs_storage/DLBS_bias_corrected/0028381_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub42512,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub42512.stripped.N4.nii.gz
+sub-CC620557,F,74.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620557.stripped.N4.nii.gz
+sub-CC120469,F,25.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120469.stripped.N4.nii.gz
+ADNI399835,F,62.3,adni_storage/ADNI_nii_gz_bias_corrected/I399835_137_S_0994.stripped.N4.nii.gz
+ADNI379704,M,60.9,adni_storage/ADNI_nii_gz_bias_corrected/I379704_131_S_5138.stripped.N4.nii.gz
+DLBS0028574_session_1,M,50.35,dlbs_storage/DLBS_bias_corrected/0028574_session_1.stripped.N4.nii.gz
+ADNI359049,M,58.9,adni_storage/ADNI_nii_gz_bias_corrected/I359049_073_S_5090.stripped.N4.nii.gz
+sub-CC222367,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222367.stripped.N4.nii.gz
+fcon1000Beijing_sub56659,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub56659.stripped.N4.nii.gz
+fcon1000Cambridge_sub63661,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub63661.stripped.N4.nii.gz
+DLBS0028566_session_1,F,57.36,dlbs_storage/DLBS_bias_corrected/0028566_session_1.stripped.N4.nii.gz
+ADNI1079630,F,57.0,adni_storage/ADNI_nii_gz_bias_corrected/I1079630_082_S_6629.stripped.N4.nii.gz
+ADNI358859,F,58.6,adni_storage/ADNI_nii_gz_bias_corrected/I358859_130_S_4982.stripped.N4.nii.gz
+DLBS0028428_session_1,F,45.85,dlbs_storage/DLBS_bias_corrected/0028428_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub99085,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub99085.stripped.N4.nii.gz
+fcon1000Beijing_sub66528,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub66528.stripped.N4.nii.gz
+ADNI1332718,F,56.0,adni_storage/ADNI_nii_gz_bias_corrected/I1332718_123_S_6884.stripped.N4.nii.gz
+ADNI1031290,F,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I1031290_070_S_6542.stripped.N4.nii.gz
+fcon1000NewYork_sub22349,M,26.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub22349.stripped.N4.nii.gz
+fcon1000Cambridge_sub06987,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub06987.stripped.N4.nii.gz
+fcon1000Beijing_sub58029,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub58029.stripped.N4.nii.gz
+sub-CC610462,F,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610462.stripped.N4.nii.gz
+ADNI1029048,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I1029048_177_S_6409.stripped.N4.nii.gz
+DLBS0028502_session_1,F,82.54,dlbs_storage/DLBS_bias_corrected/0028502_session_1.stripped.N4.nii.gz
+DLBS0028514_session_1,M,34.44,dlbs_storage/DLBS_bias_corrected/0028514_session_1.stripped.N4.nii.gz
+fcon1000Cleveland_sub34189,M,55.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub34189.stripped.N4.nii.gz
+sub-CC520517,M,65.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520517.stripped.N4.nii.gz
+fcon1000Cambridge_sub68425,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub68425.stripped.N4.nii.gz
+fcon1000Beijing_sub42555,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub42555.stripped.N4.nii.gz
+ADNI10453853,M,61.0,adni_storage/ADNI_nii_gz_bias_corrected/I10453853_019_S_10102.stripped.N4.nii.gz
+fcon1000Beijing_sub30988,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub30988.stripped.N4.nii.gz
+fcon1000Beijing_sub42843,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub42843.stripped.N4.nii.gz
+fcon1000Berlin_sub27797,M,31.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub27797.stripped.N4.nii.gz
+sub-CC610099,F,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610099.stripped.N4.nii.gz
+fcon1000Cambridge_sub37374,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub37374.stripped.N4.nii.gz
+fcon1000Cambridge_sub34586,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub34586.stripped.N4.nii.gz
+fcon1000Cambridge_sub39142,M,25.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub39142.stripped.N4.nii.gz
+ADNI1559411,F,60.2,adni_storage/ADNI_nii_gz_bias_corrected/I1559411_082_S_6629.stripped.N4.nii.gz
+fcon1000Cleveland_sub58811,M,36.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub58811.stripped.N4.nii.gz
+fcon1000Cambridge_sub25058,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub25058.stripped.N4.nii.gz
+fcon1000Cambridge_sub90059,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub90059.stripped.N4.nii.gz
+fcon1000Beijing_sub88306,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub88306.stripped.N4.nii.gz
+ADNI386097,F,58.8,adni_storage/ADNI_nii_gz_bias_corrected/I386097_041_S_4874.stripped.N4.nii.gz
+DLBS0028382_session_1,F,88.07,dlbs_storage/DLBS_bias_corrected/0028382_session_1.stripped.N4.nii.gz
+sub-CC711245,M,85.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC711245.stripped.N4.nii.gz
+ADNI908586,M,63.0,adni_storage/ADNI_nii_gz_bias_corrected/I908586_012_S_6073.stripped.N4.nii.gz
+sub-CC221977,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221977.stripped.N4.nii.gz
+fcon1000NewYork_sub48803,M,37.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub48803.stripped.N4.nii.gz
+ADNI1002373,M,58.7,adni_storage/ADNI_nii_gz_bias_corrected/I1002373_014_S_6366.stripped.N4.nii.gz
+ADNI298226,F,64.1,adni_storage/ADNI_nii_gz_bias_corrected/I298226_023_S_4448.stripped.N4.nii.gz
+fcon1000Cleveland_sub46075,M,33.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub46075.stripped.N4.nii.gz
+fcon1000NewYork_sub26267,F,15.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub26267.stripped.N4.nii.gz
+fcon1000Cambridge_sub84064,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub84064.stripped.N4.nii.gz
+fcon1000Cleveland_sub67936,F,57.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub67936.stripped.N4.nii.gz
+sub-CC520065,M,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520065.stripped.N4.nii.gz
+sub-CC320417,M,39.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320417.stripped.N4.nii.gz
+ADNI20682,M,64.5,adni_storage/ADNI_nii_gz_bias_corrected/I20682_131_S_0691.stripped.N4.nii.gz
+sub-CC310473,M,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310473.stripped.N4.nii.gz
+fcon1000NewYork_sub15758,M,32.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub15758.stripped.N4.nii.gz
+fcon1000Berlin_sub47066,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub47066.stripped.N4.nii.gz
+fcon1000Beijing_sub35309,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub35309.stripped.N4.nii.gz
+fcon1000Beijing_sub02403,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub02403.stripped.N4.nii.gz
+fcon1000Cleveland_sub18566,F,33.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub18566.stripped.N4.nii.gz
+ADNI298228,F,64.1,adni_storage/ADNI_nii_gz_bias_corrected/I298228_023_S_4448.stripped.N4.nii.gz
+sub-CC420260,F,50.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420260.stripped.N4.nii.gz
+DLBS0028550_session_1,F,22.17,dlbs_storage/DLBS_bias_corrected/0028550_session_1.stripped.N4.nii.gz
+sub-CC710350,M,81.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710350.stripped.N4.nii.gz
+fcon1000AnnArbor_sub56028,F,26.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub56028.stripped.N4.nii.gz
+fcon1000Cambridge_sub61185,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub61185.stripped.N4.nii.gz
+ADNI1592044,F,57.3,adni_storage/ADNI_nii_gz_bias_corrected/I1592044_014_S_7080.stripped.N4.nii.gz
+DLBS0028469_session_1,M,50.46,dlbs_storage/DLBS_bias_corrected/0028469_session_1.stripped.N4.nii.gz
+sub-CC510355,F,65.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510355.stripped.N4.nii.gz
+fcon1000NewYork_sub24528,M,45.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub24528.stripped.N4.nii.gz
+DLBS0028486_session_1,F,37.81,dlbs_storage/DLBS_bias_corrected/0028486_session_1.stripped.N4.nii.gz
+fcon1000Bangor_sub27519,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub27519.stripped.N4.nii.gz
+DLBS0028609_session_1,M,85.77,dlbs_storage/DLBS_bias_corrected/0028609_session_1.stripped.N4.nii.gz
+fcon1000NewYork_sub44515,M,25.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub44515.stripped.N4.nii.gz
+ADNI10907107,F,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I10907107_073_S_10167.stripped.N4.nii.gz
+fcon1000Cambridge_sub77337,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub77337.stripped.N4.nii.gz
+sub-CC420061,M,57.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420061.stripped.N4.nii.gz
+fcon1000Cambridge_sub29800,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub29800.stripped.N4.nii.gz
+sub-CC710416,F,82.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710416.stripped.N4.nii.gz
+ADNI1321252,M,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I1321252_305_S_6744.stripped.N4.nii.gz
+fcon1000Cambridge_sub61436,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub61436.stripped.N4.nii.gz
+fcon1000AnnArbor_sub04619,M,18.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub04619.stripped.N4.nii.gz
+sub-CC420433,M,51.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420433.stripped.N4.nii.gz
+fcon1000Cambridge_sub99330,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub99330.stripped.N4.nii.gz
+sub-CC420454,M,54.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420454.stripped.N4.nii.gz
+sub-CC710494,M,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710494.stripped.N4.nii.gz
+fcon1000NewYork_sub54828,M,41.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub54828.stripped.N4.nii.gz
+ADNI339298,M,63.5,adni_storage/ADNI_nii_gz_bias_corrected/I339298_073_S_4552.stripped.N4.nii.gz
+ADNI271074,M,62.6,adni_storage/ADNI_nii_gz_bias_corrected/I271074_116_S_4195.stripped.N4.nii.gz
+sub-CC720497,F,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720497.stripped.N4.nii.gz
+ADNI942628,F,62.8,adni_storage/ADNI_nii_gz_bias_corrected/I942628_036_S_6134.stripped.N4.nii.gz
+ADNI372468,F,63.5,adni_storage/ADNI_nii_gz_bias_corrected/I372468_019_S_5019.stripped.N4.nii.gz
+fcon1000Newark_sub53422,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub53422.stripped.N4.nii.gz
+sub-CC510226,M,67.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510226.stripped.N4.nii.gz
+sub-CC420241,M,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420241.stripped.N4.nii.gz
+ADNI10417928,F,56.2,adni_storage/ADNI_nii_gz_bias_corrected/I10417928_037_S_10061.stripped.N4.nii.gz
+fcon1000Beijing_sub08455,M,25.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub08455.stripped.N4.nii.gz
+sub-CC510534,M,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510534.stripped.N4.nii.gz
+fcon1000Cambridge_sub61209,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub61209.stripped.N4.nii.gz
+sub-CC721434,M,84.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721434.stripped.N4.nii.gz
+sub-CC720023,F,82.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720023.stripped.N4.nii.gz
+DLBS0028442_session_1,F,29.32,dlbs_storage/DLBS_bias_corrected/0028442_session_1.stripped.N4.nii.gz
+sub-CC410220,F,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410220.stripped.N4.nii.gz
+fcon1000Baltimore_sub31837,F,34.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub31837.stripped.N4.nii.gz
+sub-CC320680,M,41.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320680.stripped.N4.nii.gz
+DLBS0028399_session_1,F,70.68,dlbs_storage/DLBS_bias_corrected/0028399_session_1.stripped.N4.nii.gz
+ADNI1527249,F,55.8,adni_storage/ADNI_nii_gz_bias_corrected/I1527249_035_S_7030.stripped.N4.nii.gz
+sub-CC410226,M,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410226.stripped.N4.nii.gz
+sub-CC620436,M,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620436.stripped.N4.nii.gz
+fcon1000Atlanta_sub61442,F,29.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub61442.stripped.N4.nii.gz
+sub-CC610288,M,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610288.stripped.N4.nii.gz
+fcon1000Beijing_sub87089,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub87089.stripped.N4.nii.gz
+fcon1000Beijing_sub53572,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub53572.stripped.N4.nii.gz
+DLBS0028346_session_1,F,57.48,dlbs_storage/DLBS_bias_corrected/0028346_session_1.stripped.N4.nii.gz
+ADNI1463003,M,57.1,adni_storage/ADNI_nii_gz_bias_corrected/I1463003_033_S_6976.stripped.N4.nii.gz
+fcon1000Newark_sub54933,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub54933.stripped.N4.nii.gz
+sub-CC120123,F,19.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120123.stripped.N4.nii.gz
+fcon1000Beijing_sub91952,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub91952.stripped.N4.nii.gz
+DLBS0028477_session_1,F,60.81,dlbs_storage/DLBS_bias_corrected/0028477_session_1.stripped.N4.nii.gz
+DLBS0028372_session_1,F,74.83,dlbs_storage/DLBS_bias_corrected/0028372_session_1.stripped.N4.nii.gz
+sub-CC610051,M,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610051.stripped.N4.nii.gz
+fcon1000Munchen_sub66933,M,72.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub66933.stripped.N4.nii.gz
+fcon1000Atlanta_sub61902,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub61902.stripped.N4.nii.gz
+sub-CC620496,F,70.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620496.stripped.N4.nii.gz
+fcon1000Beijing_sub10973,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub10973.stripped.N4.nii.gz
+fcon1000NewYork_sub21212,M,30.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub21212.stripped.N4.nii.gz
+sub-CC520055,F,61.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520055.stripped.N4.nii.gz
+ADNI1004030,F,64.3,adni_storage/ADNI_nii_gz_bias_corrected/I1004030_023_S_6400.stripped.N4.nii.gz
+ADNI925573,M,63.4,adni_storage/ADNI_nii_gz_bias_corrected/I925573_168_S_6098.stripped.N4.nii.gz
+fcon1000Baltimore_sub81887,M,32.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub81887.stripped.N4.nii.gz
+fcon1000NewYork_sub77203,M,40.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub77203.stripped.N4.nii.gz
+sub-CC410177,M,57.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410177.stripped.N4.nii.gz
+DLBS0028338_session_1,F,30.26,dlbs_storage/DLBS_bias_corrected/0028338_session_1.stripped.N4.nii.gz
+ADNI10666918,F,59.7,adni_storage/ADNI_nii_gz_bias_corrected/I10666918_301_S_10139.stripped.N4.nii.gz
+fcon1000Cambridge_sub09015,F,25.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub09015.stripped.N4.nii.gz
+DLBS0028455_session_1,F,72.56,dlbs_storage/DLBS_bias_corrected/0028455_session_1.stripped.N4.nii.gz
+sub-CC420197,F,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420197.stripped.N4.nii.gz
+sub-CC710313,M,81.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710313.stripped.N4.nii.gz
+fcon1000Atlanta_sub55652,F,25.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub55652.stripped.N4.nii.gz
+fcon1000Leiden_sub10481,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub10481.stripped.N4.nii.gz
+fcon1000Cambridge_sub64308,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub64308.stripped.N4.nii.gz
+fcon1000Beijing_sub10869,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub10869.stripped.N4.nii.gz
+sub-CC520247,M,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520247.stripped.N4.nii.gz
+DLBS0028367_session_1,F,29.58,dlbs_storage/DLBS_bias_corrected/0028367_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub35512,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub35512.stripped.N4.nii.gz
+ADNI342679,F,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I342679_073_S_5016.stripped.N4.nii.gz
+ADNI1215081,F,56.9,adni_storage/ADNI_nii_gz_bias_corrected/I1215081_116_S_6550.stripped.N4.nii.gz
+sub-CC510304,F,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510304.stripped.N4.nii.gz
+sub-CC620549,F,73.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620549.stripped.N4.nii.gz
+DLBS0028569_session_1,M,67.65,dlbs_storage/DLBS_bias_corrected/0028569_session_1.stripped.N4.nii.gz
+sub-CC320576,F,45.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320576.stripped.N4.nii.gz
+fcon1000Cleveland_sub20003,M,39.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub20003.stripped.N4.nii.gz
+sub-CC220115,F,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220115.stripped.N4.nii.gz
+ADNI883189,F,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I883189_141_S_6042.stripped.N4.nii.gz
+ADNI415587,M,57.2,adni_storage/ADNI_nii_gz_bias_corrected/I415587_009_S_5252.stripped.N4.nii.gz
+fcon1000Atlanta_sub18702,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub18702.stripped.N4.nii.gz
+sub-CC723395,F,86.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC723395.stripped.N4.nii.gz
+fcon1000NewYork_sub53710,F,34.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub53710.stripped.N4.nii.gz
+ADNI372477,F,63.5,adni_storage/ADNI_nii_gz_bias_corrected/I372477_019_S_5019.stripped.N4.nii.gz
+ADNI238752,M,56.0,adni_storage/ADNI_nii_gz_bias_corrected/I238752_014_S_4039.stripped.N4.nii.gz
+sub-CC221031,M,38.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221031.stripped.N4.nii.gz
+fcon1000Cambridge_sub51172,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub51172.stripped.N4.nii.gz
+sub-CC620515,F,75.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620515.stripped.N4.nii.gz
+sub-CC320267,M,38.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320267.stripped.N4.nii.gz
+sub-CC221038,M,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221038.stripped.N4.nii.gz
+fcon1000AnnArbor_sub42616,F,21.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub42616.stripped.N4.nii.gz
+sub-CC321595,F,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321595.stripped.N4.nii.gz
+DLBS0028496_session_1,M,22.44,dlbs_storage/DLBS_bias_corrected/0028496_session_1.stripped.N4.nii.gz
+sub-CC720290,M,84.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720290.stripped.N4.nii.gz
+sub-CC520168,M,59.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520168.stripped.N4.nii.gz
+fcon1000Berlin_sub97162,M,28.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub97162.stripped.N4.nii.gz
+DLBS0028541_session_1,F,45.8,dlbs_storage/DLBS_bias_corrected/0028541_session_1.stripped.N4.nii.gz
+DLBS0028402_session_1,M,86.41,dlbs_storage/DLBS_bias_corrected/0028402_session_1.stripped.N4.nii.gz
+fcon1000Milwaukee_sub45019,F,44.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub45019.stripped.N4.nii.gz
+ADNI46723,M,62.9,adni_storage/ADNI_nii_gz_bias_corrected/I46723_016_S_0769.stripped.N4.nii.gz
+sub-CC720986,M,83.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720986.stripped.N4.nii.gz
+sub-CC710342,F,83.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710342.stripped.N4.nii.gz
+sub-CC520127,M,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520127.stripped.N4.nii.gz
+ADNI117520,F,61.7,adni_storage/ADNI_nii_gz_bias_corrected/I117520_051_S_1072.stripped.N4.nii.gz
+fcon1000Beijing_sub33991,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub33991.stripped.N4.nii.gz
+sub-CC420180,F,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420180.stripped.N4.nii.gz
+sub-CC610022,F,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610022.stripped.N4.nii.gz
+sub-CC521040,F,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC521040.stripped.N4.nii.gz
+fcon1000Beijing_sub62083,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub62083.stripped.N4.nii.gz
+fcon1000Atlanta_sub26938,M,31.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub26938.stripped.N4.nii.gz
+ADNI958917,F,55.6,adni_storage/ADNI_nii_gz_bias_corrected/I958917_070_S_6191.stripped.N4.nii.gz
+sub-CC420623,M,51.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420623.stripped.N4.nii.gz
+fcon1000NewYork_sub45217,M,40.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub45217.stripped.N4.nii.gz
+sub-CC310135,F,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310135.stripped.N4.nii.gz
+ADNI1427086,F,60.2,adni_storage/ADNI_nii_gz_bias_corrected/I1427086_003_S_6924.stripped.N4.nii.gz
+DLBS0028456_session_1,M,53.75,dlbs_storage/DLBS_bias_corrected/0028456_session_1.stripped.N4.nii.gz
+ADNI1549623,F,62.4,adni_storage/ADNI_nii_gz_bias_corrected/I1549623_035_S_7043.stripped.N4.nii.gz
+ADNI1485047,F,58.9,adni_storage/ADNI_nii_gz_bias_corrected/I1485047_116_S_6550.stripped.N4.nii.gz
+fcon1000Berlin_sub75506,M,28.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub75506.stripped.N4.nii.gz
+fcon1000AnnArbor_sub47659,M,33.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub47659.stripped.N4.nii.gz
+fcon1000NewYork_sub12486,M,40.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub12486.stripped.N4.nii.gz
+DLBS0028364_session_1,F,24.49,dlbs_storage/DLBS_bias_corrected/0028364_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub98624,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub98624.stripped.N4.nii.gz
+ADNI982856,F,62.6,adni_storage/ADNI_nii_gz_bias_corrected/I982856_109_S_6218.stripped.N4.nii.gz
+sub-CC321899,F,49.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321899.stripped.N4.nii.gz
+sub-CC310397,F,41.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310397.stripped.N4.nii.gz
+fcon1000NewYork_sub51677,F,25.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub51677.stripped.N4.nii.gz
+fcon1000AnnArbor_sub72215,M,20.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub72215.stripped.N4.nii.gz
+sub-CC221107,M,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221107.stripped.N4.nii.gz
+fcon1000Newark_sub76647,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub76647.stripped.N4.nii.gz
+fcon1000Atlanta_sub75153,F,32.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub75153.stripped.N4.nii.gz
+fcon1000Cambridge_sub17737,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub17737.stripped.N4.nii.gz
+ADNI296138,M,63.9,adni_storage/ADNI_nii_gz_bias_corrected/I296138_116_S_4625.stripped.N4.nii.gz
+sub-CC410094,F,54.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410094.stripped.N4.nii.gz
+fcon1000Cambridge_sub41567,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub41567.stripped.N4.nii.gz
+fcon1000Atlanta_sub24972,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub24972.stripped.N4.nii.gz
+fcon1000AnnArbor_sub90950,M,66.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub90950.stripped.N4.nii.gz
+fcon1000Beijing_sub35806,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub35806.stripped.N4.nii.gz
+sub-CC721532,F,87.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721532.stripped.N4.nii.gz
+fcon1000Cambridge_sub51512,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub51512.stripped.N4.nii.gz
+ADNI946333,F,58.4,adni_storage/ADNI_nii_gz_bias_corrected/I946333_024_S_6033.stripped.N4.nii.gz
+fcon1000Cambridge_sub71849,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub71849.stripped.N4.nii.gz
+ADNI23823,M,62.3,adni_storage/ADNI_nii_gz_bias_corrected/I23823_016_S_0769.stripped.N4.nii.gz
+DLBS0028462_session_1,M,31.03,dlbs_storage/DLBS_bias_corrected/0028462_session_1.stripped.N4.nii.gz
+fcon1000Milwaukee_sub17004,M,57.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub17004.stripped.N4.nii.gz
+ADNI10460262,F,56.4,adni_storage/ADNI_nii_gz_bias_corrected/I10460262_011_S_10041.stripped.N4.nii.gz
+sub-CC510208,F,66.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510208.stripped.N4.nii.gz
+DLBS0028524_session_1,F,56.55,dlbs_storage/DLBS_bias_corrected/0028524_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub73812,M,28.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub73812.stripped.N4.nii.gz
+sub-CC410284,F,50.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410284.stripped.N4.nii.gz
+fcon1000Atlanta_sub58250,F,54.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub58250.stripped.N4.nii.gz
+fcon1000Newark_sub71743,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub71743.stripped.N4.nii.gz
+ADNI115774,F,59.2,adni_storage/ADNI_nii_gz_bias_corrected/I115774_033_S_0733.stripped.N4.nii.gz
+ADNI394452,M,62.3,adni_storage/ADNI_nii_gz_bias_corrected/I394452_135_S_4954.stripped.N4.nii.gz
+ADNI358858,F,58.6,adni_storage/ADNI_nii_gz_bias_corrected/I358858_130_S_4982.stripped.N4.nii.gz
+sub-CC321203,M,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321203.stripped.N4.nii.gz
+fcon1000Cambridge_sub94304,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub94304.stripped.N4.nii.gz
+ADNI1591453,F,62.7,adni_storage/ADNI_nii_gz_bias_corrected/I1591453_014_S_7072.stripped.N4.nii.gz
+DLBS0028526_session_1,M,42.6,dlbs_storage/DLBS_bias_corrected/0028526_session_1.stripped.N4.nii.gz
+sub-CC310400,M,43.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310400.stripped.N4.nii.gz
+fcon1000Bangor_sub87568,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub87568.stripped.N4.nii.gz
+sub-CC520200,F,67.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520200.stripped.N4.nii.gz
+sub-CC722891,F,84.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC722891.stripped.N4.nii.gz
+sub-CC420100,M,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420100.stripped.N4.nii.gz
+sub-CC520053,M,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520053.stripped.N4.nii.gz
+fcon1000Berlin_sub23506,F,27.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub23506.stripped.N4.nii.gz
+sub-CC520287,F,59.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520287.stripped.N4.nii.gz
+DLBS0028568_session_1,M,41.17,dlbs_storage/DLBS_bias_corrected/0028568_session_1.stripped.N4.nii.gz
+ADNI1032531,F,55.9,adni_storage/ADNI_nii_gz_bias_corrected/I1032531_116_S_6550.stripped.N4.nii.gz
+sub-CC320621,F,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320621.stripped.N4.nii.gz
+fcon1000Cambridge_sub62424,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub62424.stripped.N4.nii.gz
+fcon1000Cambridge_sub43304,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub43304.stripped.N4.nii.gz
+sub-CC610061,F,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610061.stripped.N4.nii.gz
+sub-CC320478,M,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320478.stripped.N4.nii.gz
+sub-CC722421,F,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC722421.stripped.N4.nii.gz
+ADNI1469780,M,62.4,adni_storage/ADNI_nii_gz_bias_corrected/I1469780_073_S_6669.stripped.N4.nii.gz
+fcon1000Munchen_sub64202,F,67.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub64202.stripped.N4.nii.gz
+fcon1000Beijing_sub33747,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub33747.stripped.N4.nii.gz
+ADNI37255,M,62.2,adni_storage/ADNI_nii_gz_bias_corrected/I37255_067_S_1185.stripped.N4.nii.gz
+sub-CC420229,M,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420229.stripped.N4.nii.gz
+sub-CC221002,F,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221002.stripped.N4.nii.gz
+DLBS0028599_session_1,F,62.14,dlbs_storage/DLBS_bias_corrected/0028599_session_1.stripped.N4.nii.gz
+sub-CC320888,F,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320888.stripped.N4.nii.gz
+DLBS0028618_session_1,F,26.36,dlbs_storage/DLBS_bias_corrected/0028618_session_1.stripped.N4.nii.gz
+sub-CC610614,F,74.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610614.stripped.N4.nii.gz
+sub-CC710679,M,84.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710679.stripped.N4.nii.gz
+sub-CC222125,M,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222125.stripped.N4.nii.gz
+ADNI10658149,F,56.5,adni_storage/ADNI_nii_gz_bias_corrected/I10658149_016_S_10124.stripped.N4.nii.gz
+sub-CC720622,F,81.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720622.stripped.N4.nii.gz
+sub-CC712027,M,87.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC712027.stripped.N4.nii.gz
+sub-CC221487,M,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221487.stripped.N4.nii.gz
+sub-CC220352,F,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220352.stripped.N4.nii.gz
+ADNI46724,M,62.9,adni_storage/ADNI_nii_gz_bias_corrected/I46724_016_S_0769.stripped.N4.nii.gz
+fcon1000Berlin_sub57028,F,37.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub57028.stripped.N4.nii.gz
+fcon1000Beijing_sub06899,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub06899.stripped.N4.nii.gz
+fcon1000Cambridge_sub00294,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub00294.stripped.N4.nii.gz
+fcon1000Leiden_sub40907,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub40907.stripped.N4.nii.gz
+DLBS0028355_session_1,M,70.7,dlbs_storage/DLBS_bias_corrected/0028355_session_1.stripped.N4.nii.gz
+ADNI1476165,M,64.3,adni_storage/ADNI_nii_gz_bias_corrected/I1476165_305_S_6744.stripped.N4.nii.gz
+ADNI1484638,F,62.1,adni_storage/ADNI_nii_gz_bias_corrected/I1484638_067_S_6986.stripped.N4.nii.gz
+fcon1000Beijing_sub26713,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub26713.stripped.N4.nii.gz
+DLBS0028510_session_1,F,51.3,dlbs_storage/DLBS_bias_corrected/0028510_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub20563,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub20563.stripped.N4.nii.gz
+fcon1000Newark_sub36023,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub36023.stripped.N4.nii.gz
+fcon1000Berlin_sub27536,M,25.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub27536.stripped.N4.nii.gz
+fcon1000Milwaukee_sub75919,F,50.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub75919.stripped.N4.nii.gz
+sub-CC110062,M,20.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110062.stripped.N4.nii.gz
+ADNI347408,F,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I347408_130_S_4997.stripped.N4.nii.gz
+fcon1000Leiden_sub18456,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub18456.stripped.N4.nii.gz
+ADNI1571742,F,58.6,adni_storage/ADNI_nii_gz_bias_corrected/I1571742_067_S_7061.stripped.N4.nii.gz
+sub-CC610469,F,73.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610469.stripped.N4.nii.gz
+fcon1000Munchen_sub28902,M,65.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub28902.stripped.N4.nii.gz
+ADNI67599,F,58.2,adni_storage/ADNI_nii_gz_bias_corrected/I67599_033_S_0733.stripped.N4.nii.gz
+sub-CC320686,F,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320686.stripped.N4.nii.gz
+sub-CC220098,F,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220098.stripped.N4.nii.gz
+DLBS0028489_session_1,F,70.28,dlbs_storage/DLBS_bias_corrected/0028489_session_1.stripped.N4.nii.gz
+fcon1000Leiden_sub66131,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub66131.stripped.N4.nii.gz
+DLBS0028626_session_1,M,48.88,dlbs_storage/DLBS_bias_corrected/0028626_session_1.stripped.N4.nii.gz
+sub-CC720511,M,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720511.stripped.N4.nii.gz
+fcon1000Beijing_sub28403,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub28403.stripped.N4.nii.gz
+ADNI1478584,F,57.4,adni_storage/ADNI_nii_gz_bias_corrected/I1478584_037_S_6963.stripped.N4.nii.gz
+sub-CC721648,F,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721648.stripped.N4.nii.gz
+ADNI267802,F,64.1,adni_storage/ADNI_nii_gz_bias_corrected/I267802_037_S_4028.stripped.N4.nii.gz
+fcon1000Beijing_sub55856,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub55856.stripped.N4.nii.gz
+fcon1000NewYork_sub10582,M,47.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub10582.stripped.N4.nii.gz
+ADNI87683,F,57.8,adni_storage/ADNI_nii_gz_bias_corrected/I87683_133_S_0913.stripped.N4.nii.gz
+ADNI883190,F,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I883190_141_S_6042.stripped.N4.nii.gz
+fcon1000Milwaukee_sub51182,M,57.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub51182.stripped.N4.nii.gz
+sub-CC720188,M,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720188.stripped.N4.nii.gz
+sub-CC420231,M,54.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420231.stripped.N4.nii.gz
+ADNI382181,F,62.8,adni_storage/ADNI_nii_gz_bias_corrected/I382181_018_S_5240.stripped.N4.nii.gz
+DLBS0028501_session_1,F,88.36,dlbs_storage/DLBS_bias_corrected/0028501_session_1.stripped.N4.nii.gz
+DLBS0028584_session_1,F,51.24,dlbs_storage/DLBS_bias_corrected/0028584_session_1.stripped.N4.nii.gz
+ADNI336509,M,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I336509_135_S_4954.stripped.N4.nii.gz
+sub-CC610039,M,70.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610039.stripped.N4.nii.gz
+fcon1000Cambridge_sub80557,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub80557.stripped.N4.nii.gz
+fcon1000Beijing_sub77440,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub77440.stripped.N4.nii.gz
+DLBS0028532_session_1,F,34.11,dlbs_storage/DLBS_bias_corrected/0028532_session_1.stripped.N4.nii.gz
+DLBS0028505_session_1,M,20.75,dlbs_storage/DLBS_bias_corrected/0028505_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub45604,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub45604.stripped.N4.nii.gz
+sub-CC610625,F,70.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610625.stripped.N4.nii.gz
+ADNI10883387,M,60.2,adni_storage/ADNI_nii_gz_bias_corrected/I10883387_033_S_6976.stripped.N4.nii.gz
+sub-CC420435,F,53.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420435.stripped.N4.nii.gz
+fcon1000Milwaukee_sub84314,M,58.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub84314.stripped.N4.nii.gz
+DLBS0028559_session_1,F,25.01,dlbs_storage/DLBS_bias_corrected/0028559_session_1.stripped.N4.nii.gz
+fcon1000NewYork_sub27123,M,23.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub27123.stripped.N4.nii.gz
+fcon1000Beijing_sub32587,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub32587.stripped.N4.nii.gz
+DLBS0028343_session_1,F,61.27,dlbs_storage/DLBS_bias_corrected/0028343_session_1.stripped.N4.nii.gz
+sub-CC720516,F,84.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720516.stripped.N4.nii.gz
+sub-CC510086,M,65.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510086.stripped.N4.nii.gz
+fcon1000Cambridge_sub84504,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub84504.stripped.N4.nii.gz
+fcon1000Beijing_sub19642,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub19642.stripped.N4.nii.gz
+ADNI369303,F,58.8,adni_storage/ADNI_nii_gz_bias_corrected/I369303_130_S_4982.stripped.N4.nii.gz
+fcon1000Cleveland_sub17946,F,29.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub17946.stripped.N4.nii.gz
+sub-CC222555,M,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222555.stripped.N4.nii.gz
+ADNI10972872,F,64.6,adni_storage/ADNI_nii_gz_bias_corrected/I10972872_033_S_10005.stripped.N4.nii.gz
+sub-CC412004,M,54.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC412004.stripped.N4.nii.gz
+DLBS0028517_session_1,M,23.14,dlbs_storage/DLBS_bias_corrected/0028517_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub80163,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub80163.stripped.N4.nii.gz
+ADNI346548,F,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I346548_137_S_0994.stripped.N4.nii.gz
+sub-CC710664,F,82.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710664.stripped.N4.nii.gz
+fcon1000Milwaukee_sub36386,M,57.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub36386.stripped.N4.nii.gz
+fcon1000Cambridge_sub86637,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub86637.stripped.N4.nii.gz
+sub-CC420075,M,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420075.stripped.N4.nii.gz
+fcon1000Beijing_sub20246,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub20246.stripped.N4.nii.gz
+fcon1000Beijing_sub85030,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub85030.stripped.N4.nii.gz
+sub-CC320342,F,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320342.stripped.N4.nii.gz
+ADNI970943,F,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I970943_153_S_6237.stripped.N4.nii.gz
+sub-CC320759,F,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320759.stripped.N4.nii.gz
+ADNI10939738,F,62.5,adni_storage/ADNI_nii_gz_bias_corrected/I10939738_301_S_10230.stripped.N4.nii.gz
+sub-CC420091,M,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420091.stripped.N4.nii.gz
+DLBS0028611_session_1,F,46.79,dlbs_storage/DLBS_bias_corrected/0028611_session_1.stripped.N4.nii.gz
+sub-CC720180,M,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720180.stripped.N4.nii.gz
+sub-CC610496,M,70.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610496.stripped.N4.nii.gz
+sub-CC410119,M,57.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410119.stripped.N4.nii.gz
+fcon1000Beijing_sub92544,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub92544.stripped.N4.nii.gz
+sub-CC610058,M,73.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610058.stripped.N4.nii.gz
+fcon1000Bangor_sub73082,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub73082.stripped.N4.nii.gz
+sub-CC220635,F,36.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220635.stripped.N4.nii.gz
+sub-CC120987,F,20.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120987.stripped.N4.nii.gz
+sub-CC210172,F,31.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210172.stripped.N4.nii.gz
+fcon1000Milwaukee_sub56108,F,51.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub56108.stripped.N4.nii.gz
+DLBS0028330_session_1,F,35.36,dlbs_storage/DLBS_bias_corrected/0028330_session_1.stripped.N4.nii.gz
+ADNI1002815,F,64.2,adni_storage/ADNI_nii_gz_bias_corrected/I1002815_023_S_6374.stripped.N4.nii.gz
+sub-CC420137,M,56.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420137.stripped.N4.nii.gz
+DLBS0028504_session_1,F,73.07,dlbs_storage/DLBS_bias_corrected/0028504_session_1.stripped.N4.nii.gz
+sub-CC620114,M,73.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620114.stripped.N4.nii.gz
+DLBS0028591_session_1,M,86.25,dlbs_storage/DLBS_bias_corrected/0028591_session_1.stripped.N4.nii.gz
+fcon1000Leiden_sub68050,M,25.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub68050.stripped.N4.nii.gz
+DLBS0028395_session_1,M,83.87,dlbs_storage/DLBS_bias_corrected/0028395_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub13959,M,31.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub13959.stripped.N4.nii.gz
+fcon1000Cambridge_sub34741,F,27.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub34741.stripped.N4.nii.gz
+DLBS0028616_session_1,F,61.54,dlbs_storage/DLBS_bias_corrected/0028616_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub46541,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub46541.stripped.N4.nii.gz
+sub-CC120049,M,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120049.stripped.N4.nii.gz
+sub-CC610046,M,72.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610046.stripped.N4.nii.gz
+DLBS0028418_session_1,M,70.49,dlbs_storage/DLBS_bias_corrected/0028418_session_1.stripped.N4.nii.gz
+sub-CC721618,F,81.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721618.stripped.N4.nii.gz
+fcon1000Beijing_sub92490,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub92490.stripped.N4.nii.gz
+fcon1000Leiden_sub08518,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub08518.stripped.N4.nii.gz
+DLBS0028416_session_1,M,31.58,dlbs_storage/DLBS_bias_corrected/0028416_session_1.stripped.N4.nii.gz
+ADNI10883427,F,62.7,adni_storage/ADNI_nii_gz_bias_corrected/I10883427_016_S_10138.stripped.N4.nii.gz
+sub-CC121795,F,25.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121795.stripped.N4.nii.gz
+DLBS0028431_session_1,M,47.63,dlbs_storage/DLBS_bias_corrected/0028431_session_1.stripped.N4.nii.gz
+ADNI1400867,F,58.7,adni_storage/ADNI_nii_gz_bias_corrected/I1400867_019_S_6635.stripped.N4.nii.gz
+DLBS0028556_session_1,F,36.97,dlbs_storage/DLBS_bias_corrected/0028556_session_1.stripped.N4.nii.gz
+ADNI67598,F,58.2,adni_storage/ADNI_nii_gz_bias_corrected/I67598_033_S_0733.stripped.N4.nii.gz
+sub-CC321137,F,41.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321137.stripped.N4.nii.gz
+fcon1000Cambridge_sub53107,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub53107.stripped.N4.nii.gz
+fcon1000Beijing_sub98353,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub98353.stripped.N4.nii.gz
+sub-CC710131,M,85.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710131.stripped.N4.nii.gz
+DLBS0028564_session_1,F,40.21,dlbs_storage/DLBS_bias_corrected/0028564_session_1.stripped.N4.nii.gz
+ADNI86564,F,57.3,adni_storage/ADNI_nii_gz_bias_corrected/I86564_016_S_1121.stripped.N4.nii.gz
+DLBS0028461_session_1,F,53.95,dlbs_storage/DLBS_bias_corrected/0028461_session_1.stripped.N4.nii.gz
+ADNI70746,M,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I70746_016_S_0769.stripped.N4.nii.gz
+sub-CC320059,M,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320059.stripped.N4.nii.gz
+ADNI1433167,F,61.7,adni_storage/ADNI_nii_gz_bias_corrected/I1433167_024_S_6033.stripped.N4.nii.gz
+ADNI487386,M,61.1,adni_storage/ADNI_nii_gz_bias_corrected/I487386_128_S_4599.stripped.N4.nii.gz
+DLBS0028511_session_1,F,23.55,dlbs_storage/DLBS_bias_corrected/0028511_session_1.stripped.N4.nii.gz
+sub-CC610575,F,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610575.stripped.N4.nii.gz
+sub-CC610227,F,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610227.stripped.N4.nii.gz
+ADNI1031289,F,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I1031289_070_S_6542.stripped.N4.nii.gz
+fcon1000NewYork_sub63915,M,44.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub63915.stripped.N4.nii.gz
+sub-CC220828,F,33.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220828.stripped.N4.nii.gz
+DLBS0028551_session_1,M,24.66,dlbs_storage/DLBS_bias_corrected/0028551_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub08947,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub08947.stripped.N4.nii.gz
+sub-CC610050,F,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610050.stripped.N4.nii.gz
+DLBS0028433_session_1,F,80.5,dlbs_storage/DLBS_bias_corrected/0028433_session_1.stripped.N4.nii.gz
+ADNI1282405,F,57.7,adni_storage/ADNI_nii_gz_bias_corrected/I1282405_019_S_6635.stripped.N4.nii.gz
+sub-CC120816,M,23.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120816.stripped.N4.nii.gz
+sub-CC120276,F,23.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120276.stripped.N4.nii.gz
+ADNI65945,M,62.7,adni_storage/ADNI_nii_gz_bias_corrected/I65945_067_S_1185.stripped.N4.nii.gz
+fcon1000Beijing_sub55301,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub55301.stripped.N4.nii.gz
+ADNI839474,F,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I839474_141_S_6008.stripped.N4.nii.gz
+fcon1000Cambridge_sub73477,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub73477.stripped.N4.nii.gz
+fcon1000Beijing_sub12220,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub12220.stripped.N4.nii.gz
+sub-CC210422,M,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210422.stripped.N4.nii.gz
+ADNI10435128,F,55.5,adni_storage/ADNI_nii_gz_bias_corrected/I10435128_035_S_10093.stripped.N4.nii.gz
+fcon1000Cambridge_sub26348,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub26348.stripped.N4.nii.gz
diff --git a/triamese.py b/triamese.py
new file mode 100644
index 0000000..5be8c7b
--- /dev/null
+++ b/triamese.py
@@ -0,0 +1,287 @@
+import torch
+import torch.nn as nn
+import nibabel as nib
+import numpy as np
+import matplotlib.pyplot as plt
+from torch.utils.data import Dataset, DataLoader
+import os
+import pandas as pd
+from sklearn.model_selection import train_test_split
+from tqdm import tqdm
+from scipy.ndimage import zoom 
+
+def resample_nifti(img_data, target_slices = [64, 64, 32]):
+    # Determine the current number of slices along the z-axis (3rd dimension)
+    current_slices =  [img_data.shape[0], img_data.shape[1],  img_data.shape[2]]
+    # Calculate the zoom factor for resampling (only along the z-axis)
+    zoom_factor = [a / b for a, b in zip(target_slices, current_slices)]
+
+    # Resample the image data along the z-axis
+    resampled_data = zoom(img_data, (zoom_factor[0], zoom_factor[1], zoom_factor[2]), order=3)  # order=3 for cubic interpolation
+    # Ensure that the resampled data has the target number of slices
+    # print (resampled_data.shape)
+    # resampled_data = resampled_data[:target_slices,:,:]
+    # print (resampled_data.shape)
+    return resampled_data
+
+
+
+# Load metadata
+csv_path = "adni_storage/adni_brainrotnet_metadata.csv"
+df = pd.read_csv(csv_path)
+
+# Append full file paths for .nii.gz images
+data_dir = "roi"
+
+def get_image_path(row):
+    image_title = f"I{row['ImageID']}_{row['SubjectID']}.stripped.N4_cropped.nii.gz"
+    return os.path.join(data_dir, image_title)
+
+df['filepath'] = df.apply(get_image_path, axis=1)
+
+# Drop rows where the filepath doesn't exist
+df = df[df['filepath'].apply(os.path.exists)]
+
+print(df.head())  # Verify file paths and Age column
+
+
+def load_and_preprocess(filepath, patch_size=(7, 7)):
+    """
+    Load the .nii.gz file, resample it, extract three views, and divide them into fixed-size patches.
+    The number of patches is dynamically calculated based on the patch size and image dimensions.
+    """
+    import nibabel as nib
+    import torch
+
+    # Load the .nii.gz image
+    image = nib.load(filepath).get_fdata()
+    image = resample_nifti(image, [91, 109, 91])
+    
+    # Extract the three views
+    axial = image[:, :, image.shape[2] // 2]       # Axial view
+    coronal = image[:, image.shape[1] // 2, :]     # Coronal view
+    sagittal = image[image.shape[0] // 2, :, :]    # Sagittal view
+
+    def extract_patches(view, patch_size):
+        """
+        Extract fixed-size patches from a 2D view and pad to handle dimensions not divisible by patch size.
+        Dynamically calculate the number of patches.
+        """
+        h, w = view.shape
+        ph, pw = patch_size
+
+        # Calculate the number of patches in height and width
+        num_patches_h = (h + ph - 1) // ph  # Ceiling division for height
+        num_patches_w = (w + pw - 1) // pw  # Ceiling division for width
+        num_patches = num_patches_h * num_patches_w
+
+        # Pad the image to ensure dimensions are divisible by the patch size
+        pad_h = (num_patches_h * ph - h)
+        pad_w = (num_patches_w * pw - w)
+        view = torch.nn.functional.pad(torch.tensor(view, dtype=torch.float32), (0, pad_w, 0, pad_h), value=0)
+
+        # Extract patches
+        patches = []
+        for i in range(0, view.shape[0], ph):
+            for j in range(0, view.shape[1], pw):
+                patch = view[i:i + ph, j:j + pw]
+                patches.append(patch.flatten())
+
+        # Convert to tensor
+        patches = torch.stack(patches)
+        return patches, num_patches
+
+    # Extract patches and calculate dynamic num_patches
+    axial_patches, num_patches_axial = extract_patches(axial, patch_size)
+    coronal_patches, num_patches_coronal = extract_patches(coronal, patch_size)
+    sagittal_patches, num_patches_sagittal = extract_patches(sagittal, patch_size)
+
+    # Verify consistency
+    assert num_patches_axial == num_patches_coronal == num_patches_sagittal, \
+        "Mismatch in the number of patches across views."
+
+    return axial_patches, coronal_patches, sagittal_patches
+
+
+def calculate_num_patches(image_size, patch_size):
+    """
+    Calculate the total number of patches for a given image size and patch size.
+
+    Args:
+        image_size (tuple): Dimensions of the image (height, width).
+        patch_size (tuple): Dimensions of each patch (height, width).
+
+    Returns:
+        int: Total number of patches.
+    """
+    h, w = image_size
+    ph, pw = patch_size
+
+    # Calculate the number of patches along each dimension
+    num_patches_h = (h + ph - 1) // ph  # Ceiling division for height
+    num_patches_w = (w + pw - 1) // pw  # Ceiling division for width
+
+    # Total patches
+    return num_patches_h * num_patches_w
+
+num_patches = calculate_num_patches((91, 109), (7, 7))
+
+
+
+class TransformerEncoder(nn.Module):
+    def __init__(self, embed_dim, num_heads, mlp_dim, dropout=0.1):
+        super().__init__()
+        self.attn = nn.MultiheadAttention(embed_dim, num_heads)
+        self.ffn = nn.Sequential(
+            nn.Linear(embed_dim, mlp_dim),
+            nn.GELU(),
+            nn.Linear(mlp_dim, embed_dim)
+        )
+        self.norm1 = nn.LayerNorm(embed_dim)
+        self.norm2 = nn.LayerNorm(embed_dim)
+        self.dropout = nn.Dropout(dropout)
+
+    def forward(self, x):
+        x = x + self.dropout(self.attn(x, x, x)[0])
+        x = self.norm1(x)
+        x = x + self.dropout(self.ffn(x))
+        x = self.norm2(x)
+        return x
+
+class TriameseViT(nn.Module):
+    def __init__(self, num_patches, embed_dim=768, num_heads=12, mlp_dim=512, depth=10):
+        super().__init__()
+        self.patch_embedding = nn.Linear(256, embed_dim)  # Flattened patch size 16x16 -> 256
+        self.pos_embedding = nn.Parameter(torch.randn(1, num_patches, embed_dim))
+
+        # Three transformer encoders
+        self.transformers = nn.ModuleList([
+            nn.Sequential(*[TransformerEncoder(embed_dim, num_heads, mlp_dim) for _ in range(depth)])
+            for _ in range(3)
+        ])
+
+        # Three MLP heads with hidden layer size 3072
+        self.mlp_heads = nn.ModuleList([
+            nn.Sequential(
+                nn.LayerNorm(embed_dim),
+                nn.Linear(embed_dim, 3072),
+                nn.ReLU(),
+                nn.Linear(3072, 1)
+            )
+            for _ in range(3)
+        ])
+
+        # Final FNN layers: 1024 -> 512 -> 256 -> 128 -> 3 -> 1
+        self.triamese_fnn = nn.Sequential(
+            nn.Linear(3, 1024),
+            nn.ReLU(),
+            nn.Linear(1024, 512),
+            nn.ReLU(),
+            nn.Linear(512, 256),
+            nn.ReLU(),
+            nn.Linear(256, 128),
+            nn.ReLU(),
+            nn.Linear(128, 3),
+            nn.ReLU(),
+            nn.Linear(3, 1)
+        )
+
+        # Initialize weights
+        self._initialize_weights()
+
+    def _initialize_weights(self):
+        for m in self.modules():
+            if isinstance(m, nn.Linear):
+                nn.init.xavier_uniform_(m.weight)
+                nn.init.constant_(m.bias, 0)
+            elif isinstance(m, nn.LayerNorm):
+                nn.init.constant_(m.bias, 0)
+                nn.init.constant_(m.weight, 1.0)
+
+    def forward(self, axial, coronal, sagittal):
+        # Patch embedding + positional encoding
+        views = [axial, coronal, sagittal]
+        outputs = []
+
+        for i, view in enumerate(views):
+            x = self.patch_embedding(view) + self.pos_embedding
+            x = self.transformers[i](x)
+            x = x.mean(dim=1)  # Global average pooling
+            outputs.append(self.mlp_heads[i](x))
+
+        combined = torch.cat(outputs, dim=1)
+        final_output = self.triamese_fnn(combined)
+        return final_output
+    
+
+# Dataset Class
+class MRIDataset(Dataset):
+    def __init__(self, df):
+        self.filepaths = df['filepath'].values
+        self.ages = df['Age'].values
+
+    def __len__(self):
+        return len(self.filepaths)
+
+    def __getitem__(self, idx):
+        filepath = self.filepaths[idx]
+        axial, coronal, sagittal = load_and_preprocess(filepath)
+        label = torch.tensor(self.ages[idx], dtype=torch.float32)
+        return axial, coronal, sagittal, label
+
+# Split data into train and validation
+train_df, val_df = train_test_split(df, test_size=0.2, random_state=42, shuffle=True)
+
+# Create datasets
+train_dataset = MRIDataset(train_df)
+val_dataset = MRIDataset(val_df)
+
+# DataLoaders
+train_loader = DataLoader(train_dataset, batch_size=4, shuffle=False)
+val_loader = DataLoader(val_dataset, batch_size=4, shuffle=False)
+
+print(f"Train size: {len(train_dataset)}, Validation size: {len(val_dataset)}")
+
+
+model = TriameseViT(num_patches=num_patches)
+optimizer = torch.optim.Adam(model.parameters(), lr=1e-4, weight_decay=1e-6, betas=(0.9, 0.999))
+criterion = nn.MSELoss()
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+model.to(device)
+
+# Training and Validation Loop
+for epoch in range(20):
+    # Training Phase
+    model.train()
+    train_loss = 0
+    train_loader_tqdm = tqdm(train_loader, desc=f"Epoch {epoch+1} [Training]", unit="batch")
+    
+    for axial, coronal, sagittal, labels in train_loader_tqdm:
+        axial, coronal, sagittal, labels = axial.to(device), coronal.to(device), sagittal.to(device), labels.to(device)
+        
+        optimizer.zero_grad()
+        outputs = model(axial, coronal, sagittal)
+        loss = criterion(outputs, labels)
+        loss.backward()
+        optimizer.step()
+        train_loss += loss.item()
+        
+        # Update tqdm with current loss
+        train_loader_tqdm.set_postfix(loss=loss.item())
+    
+    # Validation Phase
+    model.eval()
+    val_loss = 0
+    val_loader_tqdm = tqdm(val_loader, desc=f"Epoch {epoch+1} [Validation]", unit="batch")
+    
+    with torch.no_grad():
+        for axial, coronal, sagittal, labels in val_loader_tqdm:
+            axial, coronal, sagittal, labels = axial.to(device), coronal.to(device), sagittal.to(device), labels.to(device)
+            outputs = model(axial, coronal, sagittal)
+            loss = criterion(outputs, labels)
+            val_loss += loss.item()
+            
+            # Update tqdm with current loss
+            val_loader_tqdm.set_postfix(loss=loss.item())
+    
+    print(f"Epoch {epoch+1}, Train Loss: {train_loss / len(train_loader):.4f}, Val Loss: {val_loss / len(val_loader):.4f}")
\ No newline at end of file
diff --git a/val_dataset.csv b/val_dataset.csv
new file mode 100644
index 0000000..e89aeee
--- /dev/null
+++ b/val_dataset.csv
@@ -0,0 +1,448 @@
+ImageID,Sex,Age,filepath
+fcon1000Beijing_sub10277,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub10277.stripped.N4.nii.gz
+DLBS0028329_session_1,M,30.74,dlbs_storage/DLBS_bias_corrected/0028329_session_1.stripped.N4.nii.gz
+ADNI23822,M,62.3,adni_storage/ADNI_nii_gz_bias_corrected/I23822_016_S_0769.stripped.N4.nii.gz
+ADNI1184047,F,62.6,adni_storage/ADNI_nii_gz_bias_corrected/I1184047_022_S_6013.stripped.N4.nii.gz
+sub-CC220567,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220567.stripped.N4.nii.gz
+sub-CC120309,M,27.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120309.stripped.N4.nii.gz
+fcon1000Cambridge_sub43358,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub43358.stripped.N4.nii.gz
+fcon1000Cleveland_sub46739,F,60.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub46739.stripped.N4.nii.gz
+sub-CC212153,F,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC212153.stripped.N4.nii.gz
+ADNI1017005,M,59.0,adni_storage/ADNI_nii_gz_bias_corrected/I1017005_116_S_6428.stripped.N4.nii.gz
+fcon1000Baltimore_sub17017,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub17017.stripped.N4.nii.gz
+DLBS0028449_session_1,F,28.96,dlbs_storage/DLBS_bias_corrected/0028449_session_1.stripped.N4.nii.gz
+ADNI1536458,F,61.6,adni_storage/ADNI_nii_gz_bias_corrected/I1536458_016_S_7039.stripped.N4.nii.gz
+sub-CC320297,M,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320297.stripped.N4.nii.gz
+ADNI1156695,F,64.0,adni_storage/ADNI_nii_gz_bias_corrected/I1156695_023_S_6723.stripped.N4.nii.gz
+ADNI415207,F,64.0,adni_storage/ADNI_nii_gz_bias_corrected/I415207_019_S_5019.stripped.N4.nii.gz
+fcon1000Beijing_sub89238,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub89238.stripped.N4.nii.gz
+ADNI10461419,F,58.2,adni_storage/ADNI_nii_gz_bias_corrected/I10461419_011_S_10044.stripped.N4.nii.gz
+sub-CC721292,M,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721292.stripped.N4.nii.gz
+ADNI495305,F,57.3,adni_storage/ADNI_nii_gz_bias_corrected/I495305_128_S_4607.stripped.N4.nii.gz
+sub-CC220323,F,33.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220323.stripped.N4.nii.gz
+ADNI346547,F,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I346547_137_S_0994.stripped.N4.nii.gz
+fcon1000Cambridge_sub81289,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub81289.stripped.N4.nii.gz
+ADNI10460278,F,61.6,adni_storage/ADNI_nii_gz_bias_corrected/I10460278_019_S_10106.stripped.N4.nii.gz
+fcon1000AnnArbor_sub45660,M,16.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub45660.stripped.N4.nii.gz
+sub-CC721114,F,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721114.stripped.N4.nii.gz
+fcon1000Beijing_sub95755,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub95755.stripped.N4.nii.gz
+sub-CC221980,M,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221980.stripped.N4.nii.gz
+ADNI461110,F,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I461110_137_S_0994.stripped.N4.nii.gz
+sub-CC620259,F,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620259.stripped.N4.nii.gz
+ADNI1070046,M,61.4,adni_storage/ADNI_nii_gz_bias_corrected/I1070046_006_S_6610.stripped.N4.nii.gz
+ADNI37605,F,60.1,adni_storage/ADNI_nii_gz_bias_corrected/I37605_051_S_1072.stripped.N4.nii.gz
+fcon1000Cleveland_sub82518,M,30.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub82518.stripped.N4.nii.gz
+ADNI1373547,F,62.3,adni_storage/ADNI_nii_gz_bias_corrected/I1373547_041_S_6801.stripped.N4.nii.gz
+ADNI1181047,M,60.0,adni_storage/ADNI_nii_gz_bias_corrected/I1181047_116_S_6428.stripped.N4.nii.gz
+fcon1000Leiden_sub97690,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub97690.stripped.N4.nii.gz
+fcon1000Beijing_sub58332,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub58332.stripped.N4.nii.gz
+sub-CC120462,F,18.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120462.stripped.N4.nii.gz
+fcon1000Milwaukee_sub93170,F,65.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub93170.stripped.N4.nii.gz
+fcon1000Berlin_sub47791,M,31.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub47791.stripped.N4.nii.gz
+DLBS0028387_session_1,F,54.68,dlbs_storage/DLBS_bias_corrected/0028387_session_1.stripped.N4.nii.gz
+sub-CC720400,F,86.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720400.stripped.N4.nii.gz
+DLBS0028411_session_1,F,83.09,dlbs_storage/DLBS_bias_corrected/0028411_session_1.stripped.N4.nii.gz
+ADNI77627,M,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I77627_020_S_1288.stripped.N4.nii.gz
+ADNI487400,M,61.1,adni_storage/ADNI_nii_gz_bias_corrected/I487400_128_S_4599.stripped.N4.nii.gz
+fcon1000NewYork_sub47633,F,37.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub47633.stripped.N4.nii.gz
+sub-CC410251,F,53.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410251.stripped.N4.nii.gz
+sub-CC410129,M,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410129.stripped.N4.nii.gz
+sub-CC720407,M,82.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720407.stripped.N4.nii.gz
+DLBS0028603_session_1,M,35.96,dlbs_storage/DLBS_bias_corrected/0028603_session_1.stripped.N4.nii.gz
+sub-CC410289,F,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410289.stripped.N4.nii.gz
+DLBS0028519_session_1,M,64.79,dlbs_storage/DLBS_bias_corrected/0028519_session_1.stripped.N4.nii.gz
+ADNI1331165,F,57.9,adni_storage/ADNI_nii_gz_bias_corrected/I1331165_116_S_6550.stripped.N4.nii.gz
+fcon1000AnnArbor_sub59573,F,19.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub59573.stripped.N4.nii.gz
+ADNI1028373,M,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I1028373_109_S_6406.stripped.N4.nii.gz
+sub-CC320687,F,41.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320687.stripped.N4.nii.gz
+fcon1000Cambridge_sub90681,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub90681.stripped.N4.nii.gz
+ADNI369298,F,58.8,adni_storage/ADNI_nii_gz_bias_corrected/I369298_130_S_4982.stripped.N4.nii.gz
+fcon1000NewYork_sub33062,F,34.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub33062.stripped.N4.nii.gz
+fcon1000AnnArbor_sub47191,M,27.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub47191.stripped.N4.nii.gz
+ADNI47972,F,62.8,adni_storage/ADNI_nii_gz_bias_corrected/I47972_067_S_1253.stripped.N4.nii.gz
+ADNI346266,F,61.0,adni_storage/ADNI_nii_gz_bias_corrected/I346266_036_S_4894.stripped.N4.nii.gz
+sub-CC420060,M,51.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420060.stripped.N4.nii.gz
+fcon1000Beijing_sub00440,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub00440.stripped.N4.nii.gz
+sub-CC520002,F,64.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520002.stripped.N4.nii.gz
+sub-CC710858,M,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710858.stripped.N4.nii.gz
+ADNI1002374,M,58.7,adni_storage/ADNI_nii_gz_bias_corrected/I1002374_014_S_6366.stripped.N4.nii.gz
+fcon1000Atlanta_sub86323,F,31.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub86323.stripped.N4.nii.gz
+fcon1000Baltimore_sub30072,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub30072.stripped.N4.nii.gz
+fcon1000Beijing_sub01018,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub01018.stripped.N4.nii.gz
+sub-CC120470,F,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120470.stripped.N4.nii.gz
+fcon1000Cleveland_sub75398,F,43.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub75398.stripped.N4.nii.gz
+sub-CC321107,M,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321107.stripped.N4.nii.gz
+sub-CC520253,M,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520253.stripped.N4.nii.gz
+DLBS0028509_session_1,F,51.39,dlbs_storage/DLBS_bias_corrected/0028509_session_1.stripped.N4.nii.gz
+ADNI1070047,M,61.4,adni_storage/ADNI_nii_gz_bias_corrected/I1070047_006_S_6610.stripped.N4.nii.gz
+ADNI10654619,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I10654619_100_S_10035.stripped.N4.nii.gz
+fcon1000Beijing_sub37602,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub37602.stripped.N4.nii.gz
+DLBS0028409_session_1,M,70.68,dlbs_storage/DLBS_bias_corrected/0028409_session_1.stripped.N4.nii.gz
+ADNI346866,F,56.9,adni_storage/ADNI_nii_gz_bias_corrected/I346866_128_S_4607.stripped.N4.nii.gz
+fcon1000AnnArbor_sub39923,M,70.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub39923.stripped.N4.nii.gz
+sub-CC420324,F,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420324.stripped.N4.nii.gz
+fcon1000Cleveland_sub92232,F,45.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub92232.stripped.N4.nii.gz
+fcon1000Baltimore_sub90658,F,32.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub90658.stripped.N4.nii.gz
+DLBS0028634_session_1,F,38.33,dlbs_storage/DLBS_bias_corrected/0028634_session_1.stripped.N4.nii.gz
+ADNI1536326,M,62.3,adni_storage/ADNI_nii_gz_bias_corrected/I1536326_052_S_7037.stripped.N4.nii.gz
+sub-CC510393,M,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510393.stripped.N4.nii.gz
+DLBS0028639_session_1,F,74.96,dlbs_storage/DLBS_bias_corrected/0028639_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub62843,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub62843.stripped.N4.nii.gz
+fcon1000Leiden_sub01553,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub01553.stripped.N4.nii.gz
+fcon1000Leiden_sub72247,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub72247.stripped.N4.nii.gz
+DLBS0028429_session_1,F,42.63,dlbs_storage/DLBS_bias_corrected/0028429_session_1.stripped.N4.nii.gz
+fcon1000Atlanta_sub72096,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub72096.stripped.N4.nii.gz
+fcon1000Cambridge_sub89894,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub89894.stripped.N4.nii.gz
+fcon1000Cambridge_sub69287,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub69287.stripped.N4.nii.gz
+fcon1000Beijing_sub88947,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub88947.stripped.N4.nii.gz
+sub-CC320429,M,48.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320429.stripped.N4.nii.gz
+fcon1000Cambridge_sub59434,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub59434.stripped.N4.nii.gz
+fcon1000Beijing_sub55541,M,25.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub55541.stripped.N4.nii.gz
+fcon1000Cambridge_sub42146,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub42146.stripped.N4.nii.gz
+sub-CC320107,M,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320107.stripped.N4.nii.gz
+ADNI10666922,F,64.6,adni_storage/ADNI_nii_gz_bias_corrected/I10666922_100_S_10091.stripped.N4.nii.gz
+DLBS0028588_session_1,M,48.53,dlbs_storage/DLBS_bias_corrected/0028588_session_1.stripped.N4.nii.gz
+sub-CC221324,M,36.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221324.stripped.N4.nii.gz
+DLBS0028361_session_1,M,55.56,dlbs_storage/DLBS_bias_corrected/0028361_session_1.stripped.N4.nii.gz
+ADNI10976428,F,64.2,adni_storage/ADNI_nii_gz_bias_corrected/I10976428_052_S_10275.stripped.N4.nii.gz
+sub-CC310214,F,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310214.stripped.N4.nii.gz
+DLBS0028360_session_1,F,54.03,dlbs_storage/DLBS_bias_corrected/0028360_session_1.stripped.N4.nii.gz
+sub-CC210519,F,29.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210519.stripped.N4.nii.gz
+fcon1000Cambridge_sub87846,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub87846.stripped.N4.nii.gz
+ADNI1514753,F,50.5,adni_storage/ADNI_nii_gz_bias_corrected/I1514753_037_S_7011.stripped.N4.nii.gz
+ADNI10889397,M,55.9,adni_storage/ADNI_nii_gz_bias_corrected/I10889397_019_S_10152.stripped.N4.nii.gz
+DLBS0028495_session_1,F,40.56,dlbs_storage/DLBS_bias_corrected/0028495_session_1.stripped.N4.nii.gz
+sub-CC620284,F,75.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620284.stripped.N4.nii.gz
+sub-CC210617,M,34.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210617.stripped.N4.nii.gz
+sub-CC420462,F,55.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420462.stripped.N4.nii.gz
+ADNI1151022,M,56.0,adni_storage/ADNI_nii_gz_bias_corrected/I1151022_016_S_6708.stripped.N4.nii.gz
+sub-CC420322,M,49.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420322.stripped.N4.nii.gz
+ADNI36696,F,56.4,adni_storage/ADNI_nii_gz_bias_corrected/I36696_016_S_1121.stripped.N4.nii.gz
+fcon1000AnnArbor_sub98007,F,66.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub98007.stripped.N4.nii.gz
+fcon1000Beijing_sub04191,M,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub04191.stripped.N4.nii.gz
+ADNI388757,M,64.3,adni_storage/ADNI_nii_gz_bias_corrected/I388757_116_S_4195.stripped.N4.nii.gz
+sub-CC610576,F,72.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610576.stripped.N4.nii.gz
+sub-CC720358,F,83.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720358.stripped.N4.nii.gz
+sub-CC420493,F,51.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420493.stripped.N4.nii.gz
+fcon1000Cambridge_sub81562,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub81562.stripped.N4.nii.gz
+DLBS0028408_session_1,F,20.7,dlbs_storage/DLBS_bias_corrected/0028408_session_1.stripped.N4.nii.gz
+fcon1000Leiden_sub39335,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub39335.stripped.N4.nii.gz
+ADNI907712,F,58.7,adni_storage/ADNI_nii_gz_bias_corrected/I907712_014_S_6076.stripped.N4.nii.gz
+fcon1000Beijing_sub56703,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub56703.stripped.N4.nii.gz
+ADNI313318,F,56.5,adni_storage/ADNI_nii_gz_bias_corrected/I313318_128_S_4607.stripped.N4.nii.gz
+sub-CC721888,F,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721888.stripped.N4.nii.gz
+ADNI324988,F,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I324988_023_S_4448.stripped.N4.nii.gz
+fcon1000Beijing_sub83624,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub83624.stripped.N4.nii.gz
+fcon1000Beijing_sub19974,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub19974.stripped.N4.nii.gz
+ADNI228879,F,56.0,adni_storage/ADNI_nii_gz_bias_corrected/I228879_031_S_4024.stripped.N4.nii.gz
+ADNI376810,F,59.4,adni_storage/ADNI_nii_gz_bias_corrected/I376810_067_S_5205.stripped.N4.nii.gz
+fcon1000Bangor_sub63767,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub63767.stripped.N4.nii.gz
+sub-CC320448,M,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320448.stripped.N4.nii.gz
+ADNI341795,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I341795_130_S_4982.stripped.N4.nii.gz
+ADNI140028,M,62.1,adni_storage/ADNI_nii_gz_bias_corrected/I140028_020_S_1288.stripped.N4.nii.gz
+sub-CC120795,M,24.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120795.stripped.N4.nii.gz
+ADNI382180,F,62.8,adni_storage/ADNI_nii_gz_bias_corrected/I382180_018_S_5240.stripped.N4.nii.gz
+fcon1000Beijing_sub69518,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub69518.stripped.N4.nii.gz
+sub-CC420720,F,50.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420720.stripped.N4.nii.gz
+ADNI1005735,M,55.3,adni_storage/ADNI_nii_gz_bias_corrected/I1005735_003_S_6264.stripped.N4.nii.gz
+fcon1000Cambridge_sub02953,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub02953.stripped.N4.nii.gz
+sub-CC610631,F,77.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610631.stripped.N4.nii.gz
+fcon1000Cambridge_sub17584,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub17584.stripped.N4.nii.gz
+sub-CC710591,F,85.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710591.stripped.N4.nii.gz
+fcon1000AnnArbor_sub33437,F,19.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub33437.stripped.N4.nii.gz
+sub-CC711035,F,88.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC711035.stripped.N4.nii.gz
+fcon1000Milwaukee_sub50771,F,53.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub50771.stripped.N4.nii.gz
+fcon1000Cleveland_sub64706,F,57.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub64706.stripped.N4.nii.gz
+ADNI1062510,F,62.6,adni_storage/ADNI_nii_gz_bias_corrected/I1062510_019_S_6585.stripped.N4.nii.gz
+sub-CC320218,F,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320218.stripped.N4.nii.gz
+sub-CC121144,M,26.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121144.stripped.N4.nii.gz
+sub-CC320269,F,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320269.stripped.N4.nii.gz
+fcon1000Beijing_sub50972,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub50972.stripped.N4.nii.gz
+DLBS0028593_session_1,F,62.06,dlbs_storage/DLBS_bias_corrected/0028593_session_1.stripped.N4.nii.gz
+ADNI1532649,F,59.9,adni_storage/ADNI_nii_gz_bias_corrected/I1532649_067_S_7029.stripped.N4.nii.gz
+fcon1000NewYork_sub08595,M,23.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub08595.stripped.N4.nii.gz
+fcon1000Newark_sub71042,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Newark_sub71042.stripped.N4.nii.gz
+ADNI37256,M,62.2,adni_storage/ADNI_nii_gz_bias_corrected/I37256_067_S_1185.stripped.N4.nii.gz
+fcon1000Beijing_sub22890,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub22890.stripped.N4.nii.gz
+fcon1000Beijing_sub16091,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub16091.stripped.N4.nii.gz
+fcon1000NewYork_sub51309,F,27.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub51309.stripped.N4.nii.gz
+fcon1000NewYork_sub09539,F,21.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub09539.stripped.N4.nii.gz
+sub-CC321053,F,40.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321053.stripped.N4.nii.gz
+DLBS0028481_session_1,F,41.87,dlbs_storage/DLBS_bias_corrected/0028481_session_1.stripped.N4.nii.gz
+DLBS0028597_session_1,M,52.52,dlbs_storage/DLBS_bias_corrected/0028597_session_1.stripped.N4.nii.gz
+fcon1000Bangor_sub61908,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub61908.stripped.N4.nii.gz
+sub-CC410325,F,54.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410325.stripped.N4.nii.gz
+DLBS0028347_session_1,F,85.66,dlbs_storage/DLBS_bias_corrected/0028347_session_1.stripped.N4.nii.gz
+ADNI10245972,M,59.2,adni_storage/ADNI_nii_gz_bias_corrected/I10245972_033_S_6976.stripped.N4.nii.gz
+sub-CC221033,F,28.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221033.stripped.N4.nii.gz
+ADNI1221674,F,63.4,adni_storage/ADNI_nii_gz_bias_corrected/I1221674_013_S_6780.stripped.N4.nii.gz
+DLBS0028635_session_1,F,23.5,dlbs_storage/DLBS_bias_corrected/0028635_session_1.stripped.N4.nii.gz
+ADNI345563,F,63.0,adni_storage/ADNI_nii_gz_bias_corrected/I345563_019_S_5019.stripped.N4.nii.gz
+ADNI98972,M,63.9,adni_storage/ADNI_nii_gz_bias_corrected/I98972_016_S_0769.stripped.N4.nii.gz
+ADNI254248,M,62.3,adni_storage/ADNI_nii_gz_bias_corrected/I254248_116_S_4195.stripped.N4.nii.gz
+ADNI1154285,F,63.4,adni_storage/ADNI_nii_gz_bias_corrected/I1154285_011_S_6714.stripped.N4.nii.gz
+fcon1000Beijing_sub30616,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub30616.stripped.N4.nii.gz
+DLBS0028632_session_1,F,59.96,dlbs_storage/DLBS_bias_corrected/0028632_session_1.stripped.N4.nii.gz
+ADNI10461435,F,56.1,adni_storage/ADNI_nii_gz_bias_corrected/I10461435_022_S_10066.stripped.N4.nii.gz
+DLBS0028377_session_1,F,64.37,dlbs_storage/DLBS_bias_corrected/0028377_session_1.stripped.N4.nii.gz
+ADNI1433667,F,60.4,adni_storage/ADNI_nii_gz_bias_corrected/I1433667_067_S_6912.stripped.N4.nii.gz
+ADNI1669252,M,62.2,adni_storage/ADNI_nii_gz_bias_corrected/I1669252_035_S_7121.stripped.N4.nii.gz
+fcon1000Baltimore_sub52358,F,40.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub52358.stripped.N4.nii.gz
+sub-CC610071,F,69.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610071.stripped.N4.nii.gz
+sub-CC722536,F,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC722536.stripped.N4.nii.gz
+fcon1000NewYork_sub14299,F,24.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub14299.stripped.N4.nii.gz
+fcon1000Atlanta_sub81596,M,53.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub81596.stripped.N4.nii.gz
+fcon1000Beijing_sub28965,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub28965.stripped.N4.nii.gz
+sub-CC710486,F,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC710486.stripped.N4.nii.gz
+fcon1000Cambridge_sub72068,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub72068.stripped.N4.nii.gz
+ADNI1143230,F,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I1143230_023_S_6702.stripped.N4.nii.gz
+DLBS0028573_session_1,M,44.29,dlbs_storage/DLBS_bias_corrected/0028573_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub43409,M,71.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub43409.stripped.N4.nii.gz
+fcon1000Cleveland_sub97844,F,30.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub97844.stripped.N4.nii.gz
+sub-CC620572,F,78.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620572.stripped.N4.nii.gz
+ADNI10954719,F,56.2,adni_storage/ADNI_nii_gz_bias_corrected/I10954719_052_S_10168.stripped.N4.nii.gz
+ADNI10989649,M,62.2,adni_storage/ADNI_nii_gz_bias_corrected/I10989649_019_S_10263.stripped.N4.nii.gz
+ADNI308620,M,57.0,adni_storage/ADNI_nii_gz_bias_corrected/I308620_014_S_4039.stripped.N4.nii.gz
+sub-CC721704,F,82.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721704.stripped.N4.nii.gz
+sub-CC510568,M,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510568.stripped.N4.nii.gz
+fcon1000Beijing_sub94536,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub94536.stripped.N4.nii.gz
+ADNI10392424,F,64.5,adni_storage/ADNI_nii_gz_bias_corrected/I10392424_035_S_7043.stripped.N4.nii.gz
+fcon1000Leiden_sub64642,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub64642.stripped.N4.nii.gz
+fcon1000AnnArbor_sub07921,F,19.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub07921.stripped.N4.nii.gz
+DLBS0028557_session_1,F,75.89,dlbs_storage/DLBS_bias_corrected/0028557_session_1.stripped.N4.nii.gz
+DLBS0028443_session_1,M,36.18,dlbs_storage/DLBS_bias_corrected/0028443_session_1.stripped.N4.nii.gz
+sub-CC721418,F,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721418.stripped.N4.nii.gz
+sub-CC221740,F,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221740.stripped.N4.nii.gz
+fcon1000Bangor_sub14388,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub14388.stripped.N4.nii.gz
+ADNI73491,F,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I73491_067_S_1253.stripped.N4.nii.gz
+sub-CC120218,F,27.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120218.stripped.N4.nii.gz
+fcon1000Beijing_sub28792,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub28792.stripped.N4.nii.gz
+fcon1000Beijing_sub17159,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub17159.stripped.N4.nii.gz
+ADNI1261605,F,63.3,adni_storage/ADNI_nii_gz_bias_corrected/I1261605_016_S_6802.stripped.N4.nii.gz
+fcon1000Milwaukee_sub99479,F,65.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub99479.stripped.N4.nii.gz
+sub-CC320089,F,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320089.stripped.N4.nii.gz
+fcon1000AnnArbor_sub53959,M,20.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub53959.stripped.N4.nii.gz
+fcon1000Cambridge_sub25099,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub25099.stripped.N4.nii.gz
+fcon1000Cleveland_sub65858,F,53.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub65858.stripped.N4.nii.gz
+fcon1000Cambridge_sub10268,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub10268.stripped.N4.nii.gz
+fcon1000Beijing_sub22661,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub22661.stripped.N4.nii.gz
+sub-CC620413,M,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620413.stripped.N4.nii.gz
+ADNI1634180,F,58.0,adni_storage/ADNI_nii_gz_bias_corrected/I1634180_013_S_7103.stripped.N4.nii.gz
+ADNI44673,M,60.1,adni_storage/ADNI_nii_gz_bias_corrected/I44673_020_S_1288.stripped.N4.nii.gz
+fcon1000Beijing_sub48563,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub48563.stripped.N4.nii.gz
+sub-CC410387,M,57.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC410387.stripped.N4.nii.gz
+fcon1000Beijing_sub89941,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub89941.stripped.N4.nii.gz
+DLBS0028630_session_1,M,38.37,dlbs_storage/DLBS_bias_corrected/0028630_session_1.stripped.N4.nii.gz
+sub-CC620821,M,70.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620821.stripped.N4.nii.gz
+ADNI10914005,F,59.1,adni_storage/ADNI_nii_gz_bias_corrected/I10914005_052_S_10185.stripped.N4.nii.gz
+ADNI1494601,F,58.7,adni_storage/ADNI_nii_gz_bias_corrected/I1494601_037_S_6995.stripped.N4.nii.gz
+fcon1000Cambridge_sub53296,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub53296.stripped.N4.nii.gz
+ADNI399834,F,62.3,adni_storage/ADNI_nii_gz_bias_corrected/I399834_137_S_0994.stripped.N4.nii.gz
+sub-CC420402,M,49.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420402.stripped.N4.nii.gz
+ADNI73381,F,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I73381_051_S_1072.stripped.N4.nii.gz
+fcon1000NewYork_sub10011,F,40.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub10011.stripped.N4.nii.gz
+fcon1000Atlanta_sub60499,M,29.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub60499.stripped.N4.nii.gz
+ADNI1584200,F,62.8,adni_storage/ADNI_nii_gz_bias_corrected/I1584200_035_S_7076.stripped.N4.nii.gz
+sub-CC720646,M,81.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720646.stripped.N4.nii.gz
+sub-CC221775,F,32.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221775.stripped.N4.nii.gz
+ADNI31766,F,64.6,adni_storage/ADNI_nii_gz_bias_corrected/I31766_023_S_1104.stripped.N4.nii.gz
+ADNI291512,M,62.9,adni_storage/ADNI_nii_gz_bias_corrected/I291512_116_S_4195.stripped.N4.nii.gz
+fcon1000Cambridge_sub11388,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub11388.stripped.N4.nii.gz
+DLBS0028437_session_1,F,64.01,dlbs_storage/DLBS_bias_corrected/0028437_session_1.stripped.N4.nii.gz
+DLBS0028533_session_1,F,46.77,dlbs_storage/DLBS_bias_corrected/0028533_session_1.stripped.N4.nii.gz
+ADNI974673,M,56.2,adni_storage/ADNI_nii_gz_bias_corrected/I974673_023_S_6270.stripped.N4.nii.gz
+fcon1000Cambridge_sub15258,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub15258.stripped.N4.nii.gz
+ADNI367181,M,61.8,adni_storage/ADNI_nii_gz_bias_corrected/I367181_135_S_4954.stripped.N4.nii.gz
+ADNI879215,M,56.0,adni_storage/ADNI_nii_gz_bias_corrected/I879215_114_S_6039.stripped.N4.nii.gz
+fcon1000NewYork_sub04856,M,9.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub04856.stripped.N4.nii.gz
+ADNI1604933,F,58.9,adni_storage/ADNI_nii_gz_bias_corrected/I1604933_067_S_7094.stripped.N4.nii.gz
+fcon1000Cambridge_sub62732,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub62732.stripped.N4.nii.gz
+sub-CC222652,M,30.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222652.stripped.N4.nii.gz
+sub-CC620405,F,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620405.stripped.N4.nii.gz
+sub-CC520197,F,59.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520197.stripped.N4.nii.gz
+fcon1000Cambridge_sub66351,F,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub66351.stripped.N4.nii.gz
+fcon1000Beijing_sub36942,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub36942.stripped.N4.nii.gz
+sub-CC520868,M,67.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520868.stripped.N4.nii.gz
+fcon1000AnnArbor_sub28433,M,21.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub28433.stripped.N4.nii.gz
+ADNI394453,M,62.3,adni_storage/ADNI_nii_gz_bias_corrected/I394453_135_S_4954.stripped.N4.nii.gz
+ADNI345558,F,63.0,adni_storage/ADNI_nii_gz_bias_corrected/I345558_019_S_5019.stripped.N4.nii.gz
+ADNI114057,F,57.9,adni_storage/ADNI_nii_gz_bias_corrected/I114057_016_S_1121.stripped.N4.nii.gz
+fcon1000Bangor_sub77520,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub77520.stripped.N4.nii.gz
+ADNI389098,F,59.6,adni_storage/ADNI_nii_gz_bias_corrected/I389098_067_S_5205.stripped.N4.nii.gz
+fcon1000Cambridge_sub27796,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub27796.stripped.N4.nii.gz
+fcon1000Cambridge_sub09633,F,30.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub09633.stripped.N4.nii.gz
+ADNI99328,M,61.1,adni_storage/ADNI_nii_gz_bias_corrected/I99328_020_S_1288.stripped.N4.nii.gz
+fcon1000Beijing_sub62438,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub62438.stripped.N4.nii.gz
+DLBS0028548_session_1,F,40.83,dlbs_storage/DLBS_bias_corrected/0028548_session_1.stripped.N4.nii.gz
+sub-CC420348,F,57.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420348.stripped.N4.nii.gz
+fcon1000NewYork_sub18638,M,23.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub18638.stripped.N4.nii.gz
+ADNI379689,M,60.9,adni_storage/ADNI_nii_gz_bias_corrected/I379689_131_S_5138.stripped.N4.nii.gz
+ADNI376818,F,59.4,adni_storage/ADNI_nii_gz_bias_corrected/I376818_067_S_5205.stripped.N4.nii.gz
+ADNI10443906,M,64.5,adni_storage/ADNI_nii_gz_bias_corrected/I10443906_389_S_10076.stripped.N4.nii.gz
+fcon1000Cambridge_sub41814,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub41814.stripped.N4.nii.gz
+sub-CC620451,M,75.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620451.stripped.N4.nii.gz
+ADNI705596,F,60.4,adni_storage/ADNI_nii_gz_bias_corrected/I705596_128_S_4607.stripped.N4.nii.gz
+sub-CC320850,F,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320850.stripped.N4.nii.gz
+sub-CC321428,F,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321428.stripped.N4.nii.gz
+sub-CC420071,F,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420071.stripped.N4.nii.gz
+ADNI335137,M,63.8,adni_storage/ADNI_nii_gz_bias_corrected/I335137_073_S_4552.stripped.N4.nii.gz
+ADNI10932456,M,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I10932456_035_S_7121.stripped.N4.nii.gz
+fcon1000Milwaukee_sub73547,F,55.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub73547.stripped.N4.nii.gz
+ADNI1356513,F,64.6,adni_storage/ADNI_nii_gz_bias_corrected/I1356513_137_S_6883.stripped.N4.nii.gz
+ADNI1228298,F,61.9,adni_storage/ADNI_nii_gz_bias_corrected/I1228298_131_S_6805.stripped.N4.nii.gz
+fcon1000Cambridge_sub92393,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub92393.stripped.N4.nii.gz
+fcon1000Beijing_sub18960,M,23.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub18960.stripped.N4.nii.gz
+fcon1000NewYork_sub14465,M,47.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub14465.stripped.N4.nii.gz
+DLBS0028577_session_1,M,65.52,dlbs_storage/DLBS_bias_corrected/0028577_session_1.stripped.N4.nii.gz
+sub-CC310256,M,44.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310256.stripped.N4.nii.gz
+fcon1000Beijing_sub59347,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub59347.stripped.N4.nii.gz
+DLBS0028500_session_1,M,32.85,dlbs_storage/DLBS_bias_corrected/0028500_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub69696,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub69696.stripped.N4.nii.gz
+fcon1000Milwaukee_sub23607,F,48.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub23607.stripped.N4.nii.gz
+ADNI313538,M,64.1,adni_storage/ADNI_nii_gz_bias_corrected/I313538_116_S_4625.stripped.N4.nii.gz
+fcon1000NewYork_sub03951,M,37.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub03951.stripped.N4.nii.gz
+sub-CC310407,F,39.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310407.stripped.N4.nii.gz
+ADNI1260254,M,62.4,adni_storage/ADNI_nii_gz_bias_corrected/I1260254_006_S_6610.stripped.N4.nii.gz
+ADNI114262,F,58.3,adni_storage/ADNI_nii_gz_bias_corrected/I114262_133_S_0913.stripped.N4.nii.gz
+ADNI365441,F,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I365441_073_S_5023.stripped.N4.nii.gz
+DLBS0028628_session_1,M,51.81,dlbs_storage/DLBS_bias_corrected/0028628_session_1.stripped.N4.nii.gz
+ADNI705606,F,60.4,adni_storage/ADNI_nii_gz_bias_corrected/I705606_128_S_4607.stripped.N4.nii.gz
+fcon1000Bangor_sub61418,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Bangor_sub61418.stripped.N4.nii.gz
+DLBS0028623_session_1,F,30.31,dlbs_storage/DLBS_bias_corrected/0028623_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub47278,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub47278.stripped.N4.nii.gz
+sub-CC110056,F,22.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110056.stripped.N4.nii.gz
+fcon1000Cambridge_sub47231,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub47231.stripped.N4.nii.gz
+sub-CC510648,F,61.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510648.stripped.N4.nii.gz
+sub-CC210051,F,35.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC210051.stripped.N4.nii.gz
+fcon1000Berlin_sub40143,M,34.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub40143.stripped.N4.nii.gz
+ADNI367565,M,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I367565_131_S_5138.stripped.N4.nii.gz
+DLBS0028439_session_1,M,25.38,dlbs_storage/DLBS_bias_corrected/0028439_session_1.stripped.N4.nii.gz
+fcon1000Beijing_sub55736,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub55736.stripped.N4.nii.gz
+ADNI10908346,M,60.1,adni_storage/ADNI_nii_gz_bias_corrected/I10908346_073_S_6925.stripped.N4.nii.gz
+DLBS0028615_session_1,F,80.95,dlbs_storage/DLBS_bias_corrected/0028615_session_1.stripped.N4.nii.gz
+DLBS0028326_session_1,M,66.42,dlbs_storage/DLBS_bias_corrected/0028326_session_1.stripped.N4.nii.gz
+ADNI31765,F,64.6,adni_storage/ADNI_nii_gz_bias_corrected/I31765_023_S_1104.stripped.N4.nii.gz
+DLBS0028553_session_1,M,30.76,dlbs_storage/DLBS_bias_corrected/0028553_session_1.stripped.N4.nii.gz
+ADNI1423364,F,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I1423364_003_S_6915.stripped.N4.nii.gz
+DLBS0028470_session_1,F,43.19,dlbs_storage/DLBS_bias_corrected/0028470_session_1.stripped.N4.nii.gz
+fcon1000AnnArbor_sub73168,M,78.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub73168.stripped.N4.nii.gz
+sub-CC310331,M,46.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310331.stripped.N4.nii.gz
+ADNI10952478,F,55.6,adni_storage/ADNI_nii_gz_bias_corrected/I10952478_022_S_10186.stripped.N4.nii.gz
+ADNI1032530,F,55.9,adni_storage/ADNI_nii_gz_bias_corrected/I1032530_116_S_6550.stripped.N4.nii.gz
+ADNI10939739,F,62.5,adni_storage/ADNI_nii_gz_bias_corrected/I10939739_301_S_10230.stripped.N4.nii.gz
+fcon1000Cambridge_sub88445,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub88445.stripped.N4.nii.gz
+DLBS0028582_session_1,M,48.45,dlbs_storage/DLBS_bias_corrected/0028582_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub09397,M,30.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub09397.stripped.N4.nii.gz
+sub-CC222264,M,37.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222264.stripped.N4.nii.gz
+fcon1000Leiden_sub19281,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub19281.stripped.N4.nii.gz
+ADNI1017357,F,59.0,adni_storage/ADNI_nii_gz_bias_corrected/I1017357_024_S_6033.stripped.N4.nii.gz
+fcon1000Baltimore_sub86414,F,23.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub86414.stripped.N4.nii.gz
+fcon1000Beijing_sub72678,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub72678.stripped.N4.nii.gz
+fcon1000Berlin_sub06204,M,34.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub06204.stripped.N4.nii.gz
+ADNI335031,M,63.4,adni_storage/ADNI_nii_gz_bias_corrected/I335031_116_S_4195.stripped.N4.nii.gz
+sub-CC510548,M,62.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510548.stripped.N4.nii.gz
+sub-CC220394,F,33.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC220394.stripped.N4.nii.gz
+DLBS0028523_session_1,M,49.68,dlbs_storage/DLBS_bias_corrected/0028523_session_1.stripped.N4.nii.gz
+fcon1000NewYork_sub35262,F,28.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub35262.stripped.N4.nii.gz
+sub-CC520584,F,59.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520584.stripped.N4.nii.gz
+fcon1000Beijing_sub43290,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub43290.stripped.N4.nii.gz
+fcon1000Cambridge_sub39053,M,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub39053.stripped.N4.nii.gz
+DLBS0028388_session_1,M,82.93,dlbs_storage/DLBS_bias_corrected/0028388_session_1.stripped.N4.nii.gz
+sub-CC320651,M,42.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320651.stripped.N4.nii.gz
+fcon1000Beijing_sub95575,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub95575.stripped.N4.nii.gz
+fcon1000Cambridge_sub84256,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub84256.stripped.N4.nii.gz
+sub-CC520377,F,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520377.stripped.N4.nii.gz
+fcon1000NewYork_sub30623,F,26.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub30623.stripped.N4.nii.gz
+fcon1000Beijing_sub80551,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub80551.stripped.N4.nii.gz
+sub-CC510043,M,58.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510043.stripped.N4.nii.gz
+sub-CC720238,F,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720238.stripped.N4.nii.gz
+fcon1000Baltimore_sub85922,M,30.0,fcon1000_storage/fcon1000_bias_corrected/Baltimore_sub85922.stripped.N4.nii.gz
+fcon1000Munchen_sub50162,M,73.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub50162.stripped.N4.nii.gz
+DLBS0028518_session_1,F,71.09,dlbs_storage/DLBS_bias_corrected/0028518_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub24757,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub24757.stripped.N4.nii.gz
+fcon1000Beijing_sub46259,M,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub46259.stripped.N4.nii.gz
+sub-CC420589,M,52.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420589.stripped.N4.nii.gz
+ADNI37607,F,60.1,adni_storage/ADNI_nii_gz_bias_corrected/I37607_051_S_1072.stripped.N4.nii.gz
+fcon1000NewYork_sub44979,F,15.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub44979.stripped.N4.nii.gz
+sub-CC620314,F,74.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620314.stripped.N4.nii.gz
+fcon1000Cambridge_sub55660,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub55660.stripped.N4.nii.gz
+sub-CC320359,M,47.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320359.stripped.N4.nii.gz
+sub-CC320608,M,45.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320608.stripped.N4.nii.gz
+DLBS0028595_session_1,F,87.68,dlbs_storage/DLBS_bias_corrected/0028595_session_1.stripped.N4.nii.gz
+sub-CC620193,M,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620193.stripped.N4.nii.gz
+fcon1000Cambridge_sub51671,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub51671.stripped.N4.nii.gz
+sub-CC721374,M,86.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC721374.stripped.N4.nii.gz
+fcon1000Cambridge_sub81524,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub81524.stripped.N4.nii.gz
+sub-CC221040,M,36.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC221040.stripped.N4.nii.gz
+sub-CC310361,F,43.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC310361.stripped.N4.nii.gz
+fcon1000Cambridge_sub45344,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub45344.stripped.N4.nii.gz
+ADNI1484664,F,64.4,adni_storage/ADNI_nii_gz_bias_corrected/I1484664_067_S_6525.stripped.N4.nii.gz
+DLBS0028334_session_1,F,55.26,dlbs_storage/DLBS_bias_corrected/0028334_session_1.stripped.N4.nii.gz
+ADNI10951691,F,59.0,adni_storage/ADNI_nii_gz_bias_corrected/I10951691_014_S_10236.stripped.N4.nii.gz
+fcon1000Munchen_sub74607,M,65.0,fcon1000_storage/fcon1000_bias_corrected/Munchen_sub74607.stripped.N4.nii.gz
+fcon1000Milwaukee_sub16666,F,47.0,fcon1000_storage/fcon1000_bias_corrected/Milwaukee_sub16666.stripped.N4.nii.gz
+ADNI1501367,F,56.6,adni_storage/ADNI_nii_gz_bias_corrected/I1501367_035_S_7004.stripped.N4.nii.gz
+fcon1000Cleveland_sub28596,F,24.0,fcon1000_storage/fcon1000_bias_corrected/Cleveland_sub28596.stripped.N4.nii.gz
+DLBS0028466_session_1,M,74.5,dlbs_storage/DLBS_bias_corrected/0028466_session_1.stripped.N4.nii.gz
+ADNI60862,F,57.3,adni_storage/ADNI_nii_gz_bias_corrected/I60862_133_S_0913.stripped.N4.nii.gz
+sub-CC510284,F,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510284.stripped.N4.nii.gz
+sub-CC620526,F,73.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620526.stripped.N4.nii.gz
+fcon1000Cambridge_sub64985,M,18.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub64985.stripped.N4.nii.gz
+fcon1000NewYork_sub37864,F,41.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub37864.stripped.N4.nii.gz
+sub-CC620935,M,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620935.stripped.N4.nii.gz
+sub-CC610285,M,71.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC610285.stripped.N4.nii.gz
+ADNI291506,M,62.9,adni_storage/ADNI_nii_gz_bias_corrected/I291506_116_S_4195.stripped.N4.nii.gz
+sub-CC320109,M,39.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320109.stripped.N4.nii.gz
+sub-CC620359,M,68.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620359.stripped.N4.nii.gz
+ADNI1246020,F,61.5,adni_storage/ADNI_nii_gz_bias_corrected/I1246020_003_S_6479.stripped.N4.nii.gz
+sub-CC420392,F,51.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC420392.stripped.N4.nii.gz
+sub-CC222304,M,38.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC222304.stripped.N4.nii.gz
+ADNI339295,M,63.5,adni_storage/ADNI_nii_gz_bias_corrected/I339295_073_S_4552.stripped.N4.nii.gz
+fcon1000Atlanta_sub91049,F,48.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub91049.stripped.N4.nii.gz
+DLBS0028637_session_1,F,44.54,dlbs_storage/DLBS_bias_corrected/0028637_session_1.stripped.N4.nii.gz
+ADNI1267895,M,63.2,adni_storage/ADNI_nii_gz_bias_corrected/I1267895_041_S_6136.stripped.N4.nii.gz
+DLBS0028363_session_1,M,26.02,dlbs_storage/DLBS_bias_corrected/0028363_session_1.stripped.N4.nii.gz
+fcon1000Cambridge_sub39065,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub39065.stripped.N4.nii.gz
+sub-CC510486,F,63.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC510486.stripped.N4.nii.gz
+fcon1000Cambridge_sub08588,F,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub08588.stripped.N4.nii.gz
+ADNI1092176,F,56.6,adni_storage/ADNI_nii_gz_bias_corrected/I1092176_019_S_6635.stripped.N4.nii.gz
+ADNI256318,M,56.3,adni_storage/ADNI_nii_gz_bias_corrected/I256318_014_S_4039.stripped.N4.nii.gz
+sub-CC223085,M,38.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC223085.stripped.N4.nii.gz
+fcon1000Atlanta_sub06870,M,57.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub06870.stripped.N4.nii.gz
+sub-CC720670,F,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC720670.stripped.N4.nii.gz
+DLBS0028392_session_1,F,54.85,dlbs_storage/DLBS_bias_corrected/0028392_session_1.stripped.N4.nii.gz
+fcon1000Berlin_sub28092,F,26.0,fcon1000_storage/fcon1000_bias_corrected/Berlin_sub28092.stripped.N4.nii.gz
+fcon1000AnnArbor_sub82334,M,14.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub82334.stripped.N4.nii.gz
+fcon1000Beijing_sub98617,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub98617.stripped.N4.nii.gz
+fcon1000Cambridge_sub58874,M,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub58874.stripped.N4.nii.gz
+fcon1000Beijing_sub68597,F,19.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub68597.stripped.N4.nii.gz
+ADNI90755,F,61.2,adni_storage/ADNI_nii_gz_bias_corrected/I90755_051_S_1072.stripped.N4.nii.gz
+sub-CC620264,F,76.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC620264.stripped.N4.nii.gz
+DLBS0028400_session_1,M,49.92,dlbs_storage/DLBS_bias_corrected/0028400_session_1.stripped.N4.nii.gz
+DLBS0028607_session_1,M,66.74,dlbs_storage/DLBS_bias_corrected/0028607_session_1.stripped.N4.nii.gz
+fcon1000Leiden_sub86034,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Leiden_sub86034.stripped.N4.nii.gz
+fcon1000Beijing_sub74587,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub74587.stripped.N4.nii.gz
+DLBS0028513_session_1,M,26.68,dlbs_storage/DLBS_bias_corrected/0028513_session_1.stripped.N4.nii.gz
+sub-CC320661,F,43.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC320661.stripped.N4.nii.gz
+sub-CC121479,F,26.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC121479.stripped.N4.nii.gz
+ADNI1236085,F,61.7,adni_storage/ADNI_nii_gz_bias_corrected/I1236085_022_S_6822.stripped.N4.nii.gz
+sub-CC321073,F,43.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC321073.stripped.N4.nii.gz
+DLBS0028578_session_1,M,87.34,dlbs_storage/DLBS_bias_corrected/0028578_session_1.stripped.N4.nii.gz
+ADNI871835,F,57.7,adni_storage/ADNI_nii_gz_bias_corrected/I871835_141_S_6041.stripped.N4.nii.gz
+sub-CC722651,F,79.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC722651.stripped.N4.nii.gz
+fcon1000Cambridge_sub52036,F,20.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub52036.stripped.N4.nii.gz
+ADNI228854,F,63.5,adni_storage/ADNI_nii_gz_bias_corrected/I228854_037_S_4028.stripped.N4.nii.gz
+ADNI916399,F,55.8,adni_storage/ADNI_nii_gz_bias_corrected/I916399_168_S_6085.stripped.N4.nii.gz
+ADNI982860,F,62.6,adni_storage/ADNI_nii_gz_bias_corrected/I982860_109_S_6218.stripped.N4.nii.gz
+fcon1000NewYork_sub77903,M,24.0,fcon1000_storage/fcon1000_bias_corrected/NewYork_sub77903.stripped.N4.nii.gz
+ADNI367563,M,60.7,adni_storage/ADNI_nii_gz_bias_corrected/I367563_131_S_5138.stripped.N4.nii.gz
+fcon1000Beijing_sub65467,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub65467.stripped.N4.nii.gz
+sub-CC520011,F,60.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC520011.stripped.N4.nii.gz
+fcon1000Beijing_sub72654,F,22.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub72654.stripped.N4.nii.gz
+sub-CC120008,M,26.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC120008.stripped.N4.nii.gz
+fcon1000Cambridge_sub27065,M,21.0,fcon1000_storage/fcon1000_bias_corrected/Cambridge_sub27065.stripped.N4.nii.gz
+fcon1000Beijing_sub01244,F,25.0,fcon1000_storage/fcon1000_bias_corrected/Beijing_sub01244.stripped.N4.nii.gz
+sub-CC711128,M,80.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC711128.stripped.N4.nii.gz
+sub-CC110606,M,20.0,camcan_storage/CamCAN_nii_gz_bias_corrected/sub-CC110606.stripped.N4.nii.gz
+fcon1000AnnArbor_sub97518,F,21.0,fcon1000_storage/fcon1000_bias_corrected/AnnArbor_sub97518.stripped.N4.nii.gz
+fcon1000Atlanta_sub71337,M,27.0,fcon1000_storage/fcon1000_bias_corrected/Atlanta_sub71337.stripped.N4.nii.gz
diff --git a/vit_feature_cnn_fscaled_main.py b/vit_feature_cnn_fscaled_main.py
new file mode 100644
index 0000000..56cead6
--- /dev/null
+++ b/vit_feature_cnn_fscaled_main.py
@@ -0,0 +1,423 @@
+import os
+import pandas as pd
+import nibabel as nib
+import numpy as np
+import torch
+import torch.nn as nn
+import torch.nn.init as init
+import torch.optim as optim
+from tqdm import tqdm
+from nibabel.orientations import io_orientation, axcodes2ornt, ornt_transform, apply_orientation
+from torchvision import transforms
+from transformers import ViTFeatureExtractor, ViTModel
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler
+import pickle
+import matplotlib.pyplot as plt
+import sys
+import SimpleITK as sitk
+
+def set_random_seed(seed=69420):
+    np.random.seed(seed)
+    torch.manual_seed(seed)
+    torch.cuda.manual_seed(seed)
+    torch.cuda.manual_seed_all(seed)  # For multi-GPU setups
+    torch.backends.cudnn.deterministic = True
+    torch.backends.cudnn.benchmark = False
+
+set_random_seed(69420)
+
+# Function to resample the volume to 160 slices
+def resample_volume_to_fixed_slices(data, affine, target_slices=160):
+    # Convert Numpy array and affine to SimpleITK image
+    sitk_img = sitk.GetImageFromArray(data)
+    sitk_img.SetSpacing([affine[0, 0], affine[1, 1], affine[2, 2]])
+
+    original_size = sitk_img.GetSize()  # (width, height, depth)
+    original_spacing = sitk_img.GetSpacing()  # (spacing_x, spacing_y, spacing_z)
+
+    # Calculate new spacing to achieve the target number of slices
+    new_spacing = list(original_spacing)
+    new_spacing[2] = (original_spacing[2] * original_size[2]) / target_slices
+
+    # Define new size
+    new_size = [original_size[0], original_size[1], target_slices]
+
+    # Resample the image
+    resampler = sitk.ResampleImageFilter()
+    resampler.SetOutputSpacing(new_spacing)
+    resampler.SetSize(new_size)
+    resampler.SetInterpolator(sitk.sitkLinear)
+    resampled_img = resampler.Execute(sitk_img)
+
+    return sitk.GetArrayFromImage(resampled_img)  # Return the resampled image as a numpy array
+
+# Check if GPU is available
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+# Load the CSV file into a pandas DataFrame
+csv_path = "adni_storage/adni_brainrotnet_metadata.csv"
+df = pd.read_csv(csv_path)
+# df = df.sample(n=1000, random_state=69420)
+print (df)
+# Add a new column 'filepath' with the constructed file paths
+df['filepath'] = df.apply(
+    lambda row: f"adni_storage/ADNI_nii_gz_bias_corrected/I{row['ImageID']}_{row['SubjectID']}.stripped.N4.nii.gz",
+    axis=1
+)
+
+# Load pre-trained ViT model
+feature_extractor = ViTFeatureExtractor.from_pretrained("google/vit-base-patch16-224")
+model = ViTModel.from_pretrained("google/vit-base-patch16-224")
+model.to(device)  # Move the model to the GPU (if available)
+model.eval()
+
+# Update image transform for grayscale images to match ViT input requirements
+transform = transforms.Compose([
+    transforms.ToPILImage(),
+    transforms.Resize((224, 224)),
+    transforms.Lambda(lambda img: img.convert("RGB")),  # Convert to RGB directly
+    transforms.ToTensor(),
+    transforms.Normalize(mean=feature_extractor.image_mean, std=feature_extractor.image_std),
+])
+
+# Directory to save processed images and features
+os.makedirs("adni_storage/ADNI_features", exist_ok=True)
+
+# Load pre-trained ViT model
+feature_extractor = ViTFeatureExtractor.from_pretrained("google/vit-base-patch16-224")
+model = ViTModel.from_pretrained("google/vit-base-patch16-224")
+model.to(device)  # Move the model to the GPU (if available)
+model.eval()
+
+# Update image transform for grayscale images to match ViT input requirements
+transform = transforms.Compose([
+    transforms.ToPILImage(),
+    transforms.Resize((224, 224)),
+    transforms.Lambda(lambda img: img.convert("RGB")),  # Convert to RGB directly
+    transforms.ToTensor(),
+    transforms.Normalize(mean=feature_extractor.image_mean, std=feature_extractor.image_std),
+])
+
+# Directory to save processed images and features
+os.makedirs("adni_storage/ADNI_features", exist_ok=True)
+
+# To store features and labels
+features_list = []
+labels_list = []
+
+# Process each row in the DataFrame
+for _, row in tqdm(df.iterrows(), total=len(df), desc="Processing images"):
+    filepath = row['filepath']
+    image_title = f"{row['ImageID']}_{row['SubjectID']}"
+
+    # Check if the feature file already exists
+    feature_file_path = f"adni_storage/ADNI_features/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        features = StandardScaler().fit_transform(features)
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        # print ("hiii")
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_volume_to_fixed_slices(data, affine, target_slices=160)
+
+                # Extract features for all sagittal slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+
+                    # Transform slice for ViT input
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)  # Add batch dimension and move to GPU
+
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        outputs = model(slice_tensor)
+                        slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()  # Move output back to CPU
+                        features.append(slice_features)
+
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Target is 'Age'
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, Dataset
+
+# Custom Dataset
+class ADNIDataset(Dataset):
+    def __init__(self, features_list, sex_list, age_list):
+        self.features = features_list
+        self.sex = sex_list
+        self.age = age_list
+
+    def __len__(self):
+        return len(self.features)
+
+    def __getitem__(self, idx):
+        return (
+            torch.tensor(self.features[idx], dtype=torch.float32),
+            torch.tensor(self.sex[idx], dtype=torch.float32),
+            torch.tensor(self.age[idx], dtype=torch.float32),
+        )
+
+
+# Prepare dataset and dataloaders
+sex_encoded = df['Sex'].apply(lambda x: 0 if x == 'M' else 1).tolist()
+age_list = df['Age'].tolist()
+
+# print (features_list)
+print (features_list[0].shape)
+
+# Create Dataset and DataLoader
+dataset = ADNIDataset(features_list, sex_encoded, age_list)
+train_size = int(0.8 * len(dataset))
+val_size = len(dataset) - train_size
+train_dataset, val_dataset = torch.utils.data.random_split(dataset, [train_size, val_size])
+
+# Store the indices of the validation dataset
+val_indices = val_dataset.indices
+train_indices = train_dataset.indices
+
+train_loader = DataLoader(train_dataset, batch_size=1, shuffle=False)
+val_loader = DataLoader(val_dataset, batch_size=1, shuffle=False)
+
+
+# Tracking outputs for validation samples
+val_outputs = {}
+train_outputs = {}
+
+# Initialize model, loss, and optimizer
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+
+###########################################
+# THIS IS WHERE YOU CHOOSE A MODEL TO TEST 
+###########################################
+
+import importlib
+
+# Assuming sys.argv[1] is the module name
+module_name = sys.argv[1]  # Example: "my_model"
+class_name = "AgePredictionCNN"  # The class you want to import
+
+try:
+    # Dynamically import the module
+    module = importlib.import_module(module_name)
+    
+    # Dynamically get the class
+    AgePredictionCNN = getattr(module, class_name)
+    
+    print(f"Successfully imported {class_name} from {module_name}.")
+
+except ImportError:
+    print(f"Module {module_name} could not be imported.")
+except AttributeError:
+    print(f"{class_name} does not exist in {module_name}.")
+
+##############################
+# MODEL IMPORTED DYNAMICALLY
+##############################
+
+model = AgePredictionCNN(features_list[0].shape).to(device)
+criterion = nn.L1Loss()  # MAE Loss
+optimizer = optim.Adam(model.parameters(), lr=0.001)
+best_loss = np.inf  # Initialize the best loss to infinity
+start_epoch = 0
+
+load_saved = sys.argv[2] # "last, "best"
+if load_saved != "none":
+    # Load the checkpoint
+    with open(f"model_dumps/{sys.argv[1]}_{load_saved}_model_with_metadata.pkl", "rb") as f:
+        checkpoint = pickle.load(f)
+
+    # Restore model and optimizer state
+    model.load_state_dict(checkpoint["model_state"])
+    optimizer.load_state_dict(checkpoint["optimizer_state"])
+
+    # Restore RNG states
+    torch.set_rng_state(checkpoint["t_rng_st"])
+    np.random.set_state(checkpoint["n_rng_st"])
+    if torch.cuda.is_available() and checkpoint["cuda_rng_st"] is not None:
+        torch.cuda.set_rng_state_all(checkpoint["cuda_rng_st"])
+
+    # Retrieve metadata
+    start_epoch = checkpoint["epoch"] + 1
+    loaded_loss = checkpoint["loss"]
+
+    print(f"Loaded model from epoch {start_epoch} with best validation loss: {loaded_loss:.4f}")
+
+predicted_ages = None
+# Training loop
+epochs = 200
+
+# Initialize lists to track loss
+filename = sys.argv[1] 
+csv_file = f"model_dumps/{filename}.csv"
+
+# Load existing epoch data if the file exists
+if os.path.exists(csv_file):
+    epoch_data = pd.read_csv(csv_file).to_dict(orient="records")
+    print(f"Loaded existing epoch data from {csv_file}.")
+else:
+    epoch_data = []
+    print("No existing epoch data found. Starting fresh.")
+
+
+# Plot loss vs. epoch and save the figure
+def update_loss_plot(epoch_data, filename):
+    df = pd.DataFrame(epoch_data)
+    df.to_csv(f"model_dumps/{filename}.csv", index=False)  # Save the data to CSV
+    
+    plt.figure(figsize=(8, 6))
+    plt.plot(df['epoch'], df['train_loss'], label="Train Loss", marker="o")
+    plt.plot(df['epoch'], df['val_loss'], label="Validation Loss", marker="o")
+    plt.xlabel("Epoch")
+    plt.ylabel("Loss")
+    plt.title("Loss vs. Epoch")
+    plt.legend()
+    plt.grid(True)
+    plt.savefig(f"model_dumps/{filename}.png")
+    plt.close()
+
+# Training loop
+for epoch in range(start_epoch, epochs):
+    model.train()
+    train_loss = 0.0
+    predicted_ages = []
+
+    for idx, (features, sex, age) in enumerate(train_loader):
+        features = features.unsqueeze(1).to(device)  # Add channel dimension
+        sex = sex.to(device)
+        age = age.to(device)
+        optimizer.zero_grad()
+        outputs = model(features, sex)
+        train_outputs[train_indices[idx]] = outputs.item()
+
+        loss = criterion(outputs.squeeze(), age)
+        loss.backward()
+        optimizer.step()
+
+        train_loss += loss.item()
+
+    train_loss /= len(train_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Train Loss: {train_loss:.4f}")
+
+    # Validation loop
+    model.eval()
+    val_loss = 0.0
+    with torch.no_grad():
+        for idx, (features, sex, age) in enumerate(val_loader):
+            features = features.unsqueeze(1).to(device)
+            sex = sex.to(device)
+            age = age.to(device)
+
+            outputs = model(features, sex)
+            loss = criterion(outputs.squeeze(), age)
+            val_loss += loss.item()
+
+            # Save the predicted age for the current validation sample
+            val_outputs[val_indices[idx]] = outputs.item()
+
+    val_loss /= len(val_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Validation Loss: {val_loss:.4f}")
+
+    # Save the last model with metadata
+    print(f"Saving last model...")
+    checkpoint = {
+        "model_state": model.state_dict(),
+        "optimizer_state": optimizer.state_dict(),
+        "epoch": epoch,
+        "loss": val_loss,
+        "t_rng_st": torch.get_rng_state(),
+        "n_rng_st": np.random.get_state(),
+        "cuda_rng_st": torch.cuda.get_rng_state_all() if torch.cuda.is_available() else None
+    }
+    with open(f"model_dumps/{sys.argv[1]}_last_model_with_metadata.pkl", "wb") as f:
+        pickle.dump(checkpoint, f)
+    print(f"Last model saved...")
+
+    # Check if validation loss improved
+    if val_loss < best_loss:
+        best_loss = val_loss
+        print(f"Validation loss improved to {best_loss:.4f}. Saving model...")
+        with open(f"model_dumps/{sys.argv[1]}_best_model_with_metadata.pkl", "wb") as f:
+            pickle.dump(checkpoint, f)
+
+    # Save predictions and create DataFrames (same as before)
+    # ...
+
+    # Update epoch data and save the loss plot
+    epoch_data.append({
+        "epoch": epoch + 1,
+        "train_loss": train_loss,
+        "val_loss": val_loss
+    })
+    update_loss_plot(epoch_data, sys.argv[1])
+
+    max_index = max(train_outputs.keys())
+    # Create a DataFrame with NaN for all indices initially
+    df_trn = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
+    # Assign the values to their respective indices
+    for index, value in train_outputs.items():
+        df_trn.loc[index, "Predicted_Age"] = value
+    # print (df_trn)
+
+    df2 = df.copy()
+    df2['Predicted_Age'] = df_trn['Predicted_Age']
+    train_df = df2.loc[train_outputs.keys()]
+    print (train_df)
+    train_df.to_csv(f"model_dumps/{sys.argv[1]}_predicted_ages_train.csv")
+
+    max_index = max(val_outputs.keys())
+    # Create a DataFrame with NaN for all indices initially
+    df_pred = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
+    # Assign the values to their respective indices
+    for index, value in val_outputs.items():
+        df_pred.loc[index, "Predicted_Age"] = value
+    # print (df_pred)
+
+    df1 = df.copy()
+    df1['Predicted_Age'] = df_pred['Predicted_Age']
+    test_df = df1.loc[val_outputs.keys()]
+    # print (test_df)
+    test_df.to_csv(f"model_dumps/{sys.argv[1]}_predicted_ages_val.csv")
+
+
+    # Check that the predictions have been added to the DataFrame
+    # Plot Age vs. Predicted Age
+    # plt.figure(figsize=(8, 6))
+    # plt.scatter(train_df['Age'], train_df['Predicted_Age'], color='blue', label='Predicted vs Actual')
+    # # plt.plot(test_df['Age'], test_df['Age'], color='red', linestyle='--', label='Perfect Prediction')  # Optional: Line of perfect prediction
+    # plt.xlabel('Age')
+    # plt.ylabel('Predicted Age')
+    # plt.title('Age vs Predicted Age')
+    # plt.legend()
+    # plt.grid(True)
+    # plt.show()
\ No newline at end of file
diff --git a/vit_feature_cnn_main.py b/vit_feature_cnn_main.py
index ec347cc..3987e78 100644
--- a/vit_feature_cnn_main.py
+++ b/vit_feature_cnn_main.py
@@ -386,7 +386,7 @@ def update_loss_plot(epoch_data, filename):
     # Assign the values to their respective indices
     for index, value in train_outputs.items():
         df_trn.loc[index, "Predicted_Age"] = value
-    print (df_trn)
+    # print (df_trn)
 
     df2 = df.copy()
     df2['Predicted_Age'] = df_trn['Predicted_Age']
@@ -400,12 +400,12 @@ def update_loss_plot(epoch_data, filename):
     # Assign the values to their respective indices
     for index, value in val_outputs.items():
         df_pred.loc[index, "Predicted_Age"] = value
-    print (df_pred)
+    # print (df_pred)
 
     df1 = df.copy()
     df1['Predicted_Age'] = df_pred['Predicted_Age']
     test_df = df1.loc[val_outputs.keys()]
-    print (test_df)
+    # print (test_df)
     test_df.to_csv(f"model_dumps/{sys.argv[1]}_predicted_ages_val.csv")
 
 
diff --git a/vit_feature_cnn_main1k.py b/vit_feature_cnn_main1k.py
new file mode 100644
index 0000000..0b1b803
--- /dev/null
+++ b/vit_feature_cnn_main1k.py
@@ -0,0 +1,422 @@
+import os
+import pandas as pd
+import nibabel as nib
+import numpy as np
+import torch
+import torch.nn as nn
+import torch.nn.init as init
+import torch.optim as optim
+from tqdm import tqdm
+from nibabel.orientations import io_orientation, axcodes2ornt, ornt_transform, apply_orientation
+from torchvision import transforms
+from transformers import ViTFeatureExtractor, ViTModel
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler
+import pickle
+import matplotlib.pyplot as plt
+import sys
+import SimpleITK as sitk
+
+def set_random_seed(seed=69420):
+    np.random.seed(seed)
+    torch.manual_seed(seed)
+    torch.cuda.manual_seed(seed)
+    torch.cuda.manual_seed_all(seed)  # For multi-GPU setups
+    torch.backends.cudnn.deterministic = True
+    torch.backends.cudnn.benchmark = False
+
+set_random_seed(69420)
+
+# Function to resample the volume to 160 slices
+def resample_volume_to_fixed_slices(data, affine, target_slices=160):
+    # Convert Numpy array and affine to SimpleITK image
+    sitk_img = sitk.GetImageFromArray(data)
+    sitk_img.SetSpacing([affine[0, 0], affine[1, 1], affine[2, 2]])
+
+    original_size = sitk_img.GetSize()  # (width, height, depth)
+    original_spacing = sitk_img.GetSpacing()  # (spacing_x, spacing_y, spacing_z)
+
+    # Calculate new spacing to achieve the target number of slices
+    new_spacing = list(original_spacing)
+    new_spacing[2] = (original_spacing[2] * original_size[2]) / target_slices
+
+    # Define new size
+    new_size = [original_size[0], original_size[1], target_slices]
+
+    # Resample the image
+    resampler = sitk.ResampleImageFilter()
+    resampler.SetOutputSpacing(new_spacing)
+    resampler.SetSize(new_size)
+    resampler.SetInterpolator(sitk.sitkLinear)
+    resampled_img = resampler.Execute(sitk_img)
+
+    return sitk.GetArrayFromImage(resampled_img)  # Return the resampled image as a numpy array
+
+# Check if GPU is available
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+# Load the CSV file into a pandas DataFrame
+csv_path = "adni_storage/adni_brainrotnet_metadata.csv"
+df = pd.read_csv(csv_path)
+df = df.sample(n=1000, random_state=69420)
+print (df)
+# Add a new column 'filepath' with the constructed file paths
+df['filepath'] = df.apply(
+    lambda row: f"adni_storage/ADNI_nii_gz_bias_corrected/I{row['ImageID']}_{row['SubjectID']}.stripped.N4.nii.gz",
+    axis=1
+)
+
+# Load pre-trained ViT model
+feature_extractor = ViTFeatureExtractor.from_pretrained("google/vit-base-patch16-224")
+model = ViTModel.from_pretrained("google/vit-base-patch16-224")
+model.to(device)  # Move the model to the GPU (if available)
+model.eval()
+
+# Update image transform for grayscale images to match ViT input requirements
+transform = transforms.Compose([
+    transforms.ToPILImage(),
+    transforms.Resize((224, 224)),
+    transforms.Lambda(lambda img: img.convert("RGB")),  # Convert to RGB directly
+    transforms.ToTensor(),
+    transforms.Normalize(mean=feature_extractor.image_mean, std=feature_extractor.image_std),
+])
+
+# Directory to save processed images and features
+os.makedirs("adni_storage/ADNI_features", exist_ok=True)
+
+# Load pre-trained ViT model
+feature_extractor = ViTFeatureExtractor.from_pretrained("google/vit-base-patch16-224")
+model = ViTModel.from_pretrained("google/vit-base-patch16-224")
+model.to(device)  # Move the model to the GPU (if available)
+model.eval()
+
+# Update image transform for grayscale images to match ViT input requirements
+transform = transforms.Compose([
+    transforms.ToPILImage(),
+    transforms.Resize((224, 224)),
+    transforms.Lambda(lambda img: img.convert("RGB")),  # Convert to RGB directly
+    transforms.ToTensor(),
+    transforms.Normalize(mean=feature_extractor.image_mean, std=feature_extractor.image_std),
+])
+
+# Directory to save processed images and features
+os.makedirs("adni_storage/ADNI_features", exist_ok=True)
+
+# To store features and labels
+features_list = []
+labels_list = []
+
+# Process each row in the DataFrame
+for _, row in tqdm(df.iterrows(), total=len(df), desc="Processing images"):
+    filepath = row['filepath']
+    image_title = f"{row['ImageID']}_{row['SubjectID']}"
+
+    # Check if the feature file already exists
+    feature_file_path = f"adni_storage/ADNI_features/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        # print ("hiii")
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_volume_to_fixed_slices(data, affine, target_slices=160)
+
+                # Extract features for all sagittal slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+
+                    # Transform slice for ViT input
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)  # Add batch dimension and move to GPU
+
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        outputs = model(slice_tensor)
+                        slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()  # Move output back to CPU
+                        features.append(slice_features)
+
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Target is 'Age'
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, Dataset
+
+# Custom Dataset
+class ADNIDataset(Dataset):
+    def __init__(self, features_list, sex_list, age_list):
+        self.features = features_list
+        self.sex = sex_list
+        self.age = age_list
+
+    def __len__(self):
+        return len(self.features)
+
+    def __getitem__(self, idx):
+        return (
+            torch.tensor(self.features[idx], dtype=torch.float32),
+            torch.tensor(self.sex[idx], dtype=torch.float32),
+            torch.tensor(self.age[idx], dtype=torch.float32),
+        )
+
+
+# Prepare dataset and dataloaders
+sex_encoded = df['Sex'].apply(lambda x: 0 if x == 'M' else 1).tolist()
+age_list = df['Age'].tolist()
+
+# print (features_list)
+print (features_list[0].shape)
+
+# Create Dataset and DataLoader
+dataset = ADNIDataset(features_list, sex_encoded, age_list)
+train_size = int(0.8 * len(dataset))
+val_size = len(dataset) - train_size
+train_dataset, val_dataset = torch.utils.data.random_split(dataset, [train_size, val_size])
+
+# Store the indices of the validation dataset
+val_indices = val_dataset.indices
+train_indices = train_dataset.indices
+
+train_loader = DataLoader(train_dataset, batch_size=1, shuffle=False)
+val_loader = DataLoader(val_dataset, batch_size=1, shuffle=False)
+
+
+# Tracking outputs for validation samples
+val_outputs = {}
+train_outputs = {}
+
+# Initialize model, loss, and optimizer
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+
+###########################################
+# THIS IS WHERE YOU CHOOSE A MODEL TO TEST 
+###########################################
+
+import importlib
+
+# Assuming sys.argv[1] is the module name
+module_name = sys.argv[1]  # Example: "my_model"
+class_name = "AgePredictionCNN"  # The class you want to import
+
+try:
+    # Dynamically import the module
+    module = importlib.import_module(module_name)
+    
+    # Dynamically get the class
+    AgePredictionCNN = getattr(module, class_name)
+    
+    print(f"Successfully imported {class_name} from {module_name}.")
+
+except ImportError:
+    print(f"Module {module_name} could not be imported.")
+except AttributeError:
+    print(f"{class_name} does not exist in {module_name}.")
+
+##############################
+# MODEL IMPORTED DYNAMICALLY
+##############################
+
+model = AgePredictionCNN(features_list[0].shape).to(device)
+criterion = nn.L1Loss()  # MAE Loss
+optimizer = optim.Adam(model.parameters(), lr=0.001)
+best_loss = np.inf  # Initialize the best loss to infinity
+start_epoch = 0
+
+load_saved = sys.argv[2] # "last, "best"
+if load_saved != "none":
+    # Load the checkpoint
+    with open(f"model_dumps/{sys.argv[1]}_{load_saved}_model_with_metadata_1k.pkl", "rb") as f:
+        checkpoint = pickle.load(f)
+
+    # Restore model and optimizer state
+    model.load_state_dict(checkpoint["model_state"])
+    optimizer.load_state_dict(checkpoint["optimizer_state"])
+
+    # Restore RNG states
+    torch.set_rng_state(checkpoint["t_rng_st"])
+    np.random.set_state(checkpoint["n_rng_st"])
+    if torch.cuda.is_available() and checkpoint["cuda_rng_st"] is not None:
+        torch.cuda.set_rng_state_all(checkpoint["cuda_rng_st"])
+
+    # Retrieve metadata
+    start_epoch = checkpoint["epoch"] + 1
+    loaded_loss = checkpoint["loss"]
+
+    print(f"Loaded model from epoch {start_epoch} with best validation loss: {loaded_loss:.4f}")
+
+predicted_ages = None
+# Training loop
+epochs = 200
+
+# Initialize lists to track loss
+filename = sys.argv[1] 
+csv_file = f"model_dumps/{filename}1k.csv"
+
+# Load existing epoch data if the file exists
+if os.path.exists(csv_file):
+    epoch_data = pd.read_csv(csv_file).to_dict(orient="records")
+    print(f"Loaded existing epoch data from {csv_file}.")
+else:
+    epoch_data = []
+    print("No existing epoch data found. Starting fresh.")
+
+
+# Plot loss vs. epoch and save the figure
+def update_loss_plot(epoch_data, filename):
+    df = pd.DataFrame(epoch_data)
+    df.to_csv(f"model_dumps/{filename}1k.csv", index=False)  # Save the data to CSV
+    
+    plt.figure(figsize=(8, 6))
+    plt.plot(df['epoch'], df['train_loss'], label="Train Loss", marker="o")
+    plt.plot(df['epoch'], df['val_loss'], label="Validation Loss", marker="o")
+    plt.xlabel("Epoch")
+    plt.ylabel("Loss")
+    plt.title("Loss vs. Epoch")
+    plt.legend()
+    plt.grid(True)
+    plt.savefig(f"model_dumps/{filename}1k.png")
+    plt.close()
+
+# Training loop
+for epoch in range(start_epoch, epochs):
+    model.train()
+    train_loss = 0.0
+    predicted_ages = []
+
+    for idx, (features, sex, age) in enumerate(train_loader):
+        features = features.unsqueeze(1).to(device)  # Add channel dimension
+        sex = sex.to(device)
+        age = age.to(device)
+        optimizer.zero_grad()
+        outputs = model(features, sex)
+        train_outputs[train_indices[idx]] = outputs.item()
+
+        loss = criterion(outputs.squeeze(), age)
+        loss.backward()
+        optimizer.step()
+
+        train_loss += loss.item()
+
+    train_loss /= len(train_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Train Loss: {train_loss:.4f}")
+
+    # Validation loop
+    model.eval()
+    val_loss = 0.0
+    with torch.no_grad():
+        for idx, (features, sex, age) in enumerate(val_loader):
+            features = features.unsqueeze(1).to(device)
+            sex = sex.to(device)
+            age = age.to(device)
+
+            outputs = model(features, sex)
+            loss = criterion(outputs.squeeze(), age)
+            val_loss += loss.item()
+
+            # Save the predicted age for the current validation sample
+            val_outputs[val_indices[idx]] = outputs.item()
+
+    val_loss /= len(val_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Validation Loss: {val_loss:.4f}")
+
+    # Save the last model with metadata
+    print(f"Saving last model...")
+    checkpoint = {
+        "model_state": model.state_dict(),
+        "optimizer_state": optimizer.state_dict(),
+        "epoch": epoch,
+        "loss": val_loss,
+        "t_rng_st": torch.get_rng_state(),
+        "n_rng_st": np.random.get_state(),
+        "cuda_rng_st": torch.cuda.get_rng_state_all() if torch.cuda.is_available() else None
+    }
+    with open(f"model_dumps/{sys.argv[1]}_last_model_with_metadata.pkl", "wb") as f:
+        pickle.dump(checkpoint, f)
+    print(f"Last model saved...")
+
+    # Check if validation loss improved
+    if val_loss < best_loss:
+        best_loss = val_loss
+        print(f"Validation loss improved to {best_loss:.4f}. Saving model...")
+        with open(f"model_dumps/{sys.argv[1]}_best_model_with_metadata1k.pkl", "wb") as f:
+            pickle.dump(checkpoint, f)
+
+    # Save predictions and create DataFrames (same as before)
+    # ...
+
+    # Update epoch data and save the loss plot
+    epoch_data.append({
+        "epoch": epoch + 1,
+        "train_loss": train_loss,
+        "val_loss": val_loss
+    })
+    update_loss_plot(epoch_data, sys.argv[1])
+
+    max_index = max(train_outputs.keys())
+    # Create a DataFrame with NaN for all indices initially
+    df_trn = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
+    # Assign the values to their respective indices
+    for index, value in train_outputs.items():
+        df_trn.loc[index, "Predicted_Age"] = value
+    # print (df_trn)
+
+    df2 = df.copy()
+    # df2['Predicted_Age'] = df_trn['Predicted_Age']
+    # train_df = df2.loc[train_outputs.keys()]
+    # print (train_df)
+    # train_df.to_csv(f"model_dumps/{sys.argv[1]}_predicted_ages_train1k.csv")
+
+    # max_index = max(val_outputs.keys())
+    # # Create a DataFrame with NaN for all indices initially
+    # df_pred = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
+    # # Assign the values to their respective indices
+    # for index, value in val_outputs.items():
+    #     df_pred.loc[index, "Predicted_Age"] = value
+    # # print (df_pred)
+
+    # df1 = df.copy()
+    # df1['Predicted_Age'] = df_pred['Predicted_Age']
+    # test_df = df1.loc[val_outputs.keys()]
+    # # print (test_df)
+    # test_df.to_csv(f"model_dumps/{sys.argv[1]}_predicted_ages_val1k.csv")
+
+
+    # Check that the predictions have been added to the DataFrame
+    # Plot Age vs. Predicted Age
+    # plt.figure(figsize=(8, 6))
+    # plt.scatter(train_df['Age'], train_df['Predicted_Age'], color='blue', label='Predicted vs Actual')
+    # # plt.plot(test_df['Age'], test_df['Age'], color='red', linestyle='--', label='Perfect Prediction')  # Optional: Line of perfect prediction
+    # plt.xlabel('Age')
+    # plt.ylabel('Predicted Age')
+    # plt.title('Age vs Predicted Age')
+    # plt.legend()
+    # plt.grid(True)
+    # plt.show()
\ No newline at end of file
diff --git a/vit_feature_cnn_main1kroi.py b/vit_feature_cnn_main1kroi.py
new file mode 100644
index 0000000..7e6d690
--- /dev/null
+++ b/vit_feature_cnn_main1kroi.py
@@ -0,0 +1,466 @@
+import os
+import pandas as pd
+import nibabel as nib
+import numpy as np
+import torch
+import torch.nn as nn
+import torch.nn.init as init
+import torch.optim as optim
+from tqdm import tqdm
+from nibabel.orientations import io_orientation, axcodes2ornt, ornt_transform, apply_orientation
+from torchvision import transforms
+from transformers import ViTFeatureExtractor, ViTModel
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler
+import pickle
+import matplotlib.pyplot as plt
+import sys
+import SimpleITK as sitk
+from scipy.ndimage import zoom
+
+def set_random_seed(seed=69420):
+    np.random.seed(seed)
+    torch.manual_seed(seed)
+    torch.cuda.manual_seed(seed)
+    torch.cuda.manual_seed_all(seed)  # For multi-GPU setups
+    torch.backends.cudnn.deterministic = True
+    torch.backends.cudnn.benchmark = False
+
+set_random_seed(69420)
+
+# Function to resample the volume to 160 slices
+def resample_volume_to_fixed_slices(data, affine, target_slices=160):
+    # Convert Numpy array and affine to SimpleITK image
+    sitk_img = sitk.GetImageFromArray(data)
+    sitk_img.SetSpacing([affine[0, 0], affine[1, 1], affine[2, 2]])
+
+    original_size = sitk_img.GetSize()  # (width, height, depth)
+    original_spacing = sitk_img.GetSpacing()  # (spacing_x, spacing_y, spacing_z)
+
+    # Calculate new spacing to achieve the target number of slices
+    new_spacing = list(original_spacing)
+    new_spacing[2] = (original_spacing[2] * original_size[2]) / target_slices
+
+    # Define new size
+    new_size = [original_size[0], original_size[1], target_slices]
+
+    # Resample the image
+    resampler = sitk.ResampleImageFilter()
+    resampler.SetOutputSpacing(new_spacing)
+    resampler.SetSize(new_size)
+    resampler.SetInterpolator(sitk.sitkLinear)
+    resampled_img = resampler.Execute(sitk_img)
+
+    return sitk.GetArrayFromImage(resampled_img)  # Return the resampled image as a numpy array
+
+def resample_nifti(img_data, target_slices = 160):
+    # Determine the current number of slices along the z-axis (3rd dimension)
+    current_slices = img_data.shape[0]
+    # Calculate the zoom factor for resampling (only along the z-axis)
+    zoom_factor = target_slices / current_slices
+    # Resample the image data along the z-axis
+    resampled_data = zoom(img_data, (zoom_factor, 1, 1), order=3)  # order=3 for cubic interpolation
+    # Ensure that the resampled data has the target number of slices
+    # print (resampled_data.shape)
+    # resampled_data = resampled_data[:target_slices,:,:]
+    # print (resampled_data.shape)
+    return resampled_data
+
+# Check if GPU is available
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+# Load the CSV file into a pandas DataFrame
+csv_path = "adni_storage/adni_brainrotnet_metadata.csv"
+df = pd.read_csv(csv_path)
+df = df.sample(n=1000, random_state=69420)
+print (df)
+# Add a new column 'filepath' with the constructed file paths
+df['filepath'] = df.apply(
+    lambda row: f"adni_storage/ADNI_nii_gz_bias_corrected/I{row['ImageID']}_{row['SubjectID']}.stripped.N4.nii.gz",
+    axis=1
+)
+
+# Load pre-trained ViT model
+feature_extractor = ViTFeatureExtractor.from_pretrained("google/vit-base-patch16-224")
+model = ViTModel.from_pretrained("google/vit-base-patch16-224")
+model.to(device)  # Move the model to the GPU (if available)
+model.eval()
+
+# Update image transform for grayscale images to match ViT input requirements
+transform = transforms.Compose([
+    transforms.ToPILImage(),
+    transforms.Resize((224, 224)),
+    transforms.Lambda(lambda img: img.convert("RGB")),  # Convert to RGB directly
+    transforms.ToTensor(),
+    transforms.Normalize(mean=feature_extractor.image_mean, std=feature_extractor.image_std),
+])
+
+# Directory to save processed images and features
+os.makedirs("adni_storage/ADNI_features", exist_ok=True)
+
+# Load pre-trained ViT model
+feature_extractor = ViTFeatureExtractor.from_pretrained("google/vit-base-patch16-224")
+model = ViTModel.from_pretrained("google/vit-base-patch16-224")
+model.to(device)  # Move the model to the GPU (if available)
+model.eval()
+
+# Update image transform for grayscale images to match ViT input requirements
+transform = transforms.Compose([
+    transforms.ToPILImage(),
+    transforms.Resize((224, 224)),
+    transforms.Lambda(lambda img: img.convert("RGB")),  # Convert to RGB directly
+    transforms.ToTensor(),
+    transforms.Normalize(mean=feature_extractor.image_mean, std=feature_extractor.image_std),
+])
+
+# Directory to save processed images and features
+os.makedirs("adni_storage/ADNI_features/roi", exist_ok=True)
+
+# To store features and labels
+features_list = []
+labels_list = []
+
+def calculate_bounding_box_from_volume(volume):
+    # Find indices of non-zero values
+    non_zero_indices = np.argwhere(volume > 0)
+
+    # Calculate min and max indices along each dimension
+    min_indices = np.min(non_zero_indices, axis=0)
+    max_indices = np.max(non_zero_indices, axis=0)
+
+    # Convert indices to integers
+    min_indices = min_indices.astype(int)
+    max_indices = max_indices.astype(int)
+
+    return min_indices, max_indices
+
+
+def crop_brain_volumes(brain_data):
+    
+
+        # Calculate bounding box from the brain volume
+    min_indices, max_indices = calculate_bounding_box_from_volume(brain_data)
+
+        # Crop the volume
+    cropped_brain = brain_data[min_indices[0]:max_indices[0] + 1,
+                                   min_indices[1]:max_indices[1] + 1,
+                                   min_indices[2]:max_indices[2] + 1]
+    return cropped_brain
+
+# Process each row in the DataFrame
+for _, row in tqdm(df.iterrows(), total=len(df), desc="Processing images"):
+    filepath = row['filepath']
+    image_title = f"{row['ImageID']}_{row['SubjectID']}"
+
+    # Check if the feature file already exists
+    feature_file_path = f"adni_storage/ADNI_features/roi/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        # print ("hiii")
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Crop the brain volume
+                data = crop_brain_volumes(data)
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+
+                # Extract features for all sagittal slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+
+                    # Transform slice for ViT input
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)  # Add batch dimension and move to GPU
+
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        outputs = model(slice_tensor)
+                        slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()  # Move output back to CPU
+                        features.append(slice_features)
+
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Target is 'Age'
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, Dataset
+
+# Custom Dataset
+class ADNIDataset(Dataset):
+    def __init__(self, features_list, sex_list, age_list):
+        self.features = features_list
+        self.sex = sex_list
+        self.age = age_list
+
+    def __len__(self):
+        return len(self.features)
+
+    def __getitem__(self, idx):
+        return (
+            torch.tensor(self.features[idx], dtype=torch.float32),
+            torch.tensor(self.sex[idx], dtype=torch.float32),
+            torch.tensor(self.age[idx], dtype=torch.float32),
+        )
+
+
+# Prepare dataset and dataloaders
+sex_encoded = df['Sex'].apply(lambda x: 0 if x == 'M' else 1).tolist()
+age_list = df['Age'].tolist()
+
+# print (features_list)
+print (features_list[0].shape)
+
+# Create Dataset and DataLoader
+dataset = ADNIDataset(features_list, sex_encoded, age_list)
+train_size = int(0.8 * len(dataset))
+val_size = len(dataset) - train_size
+train_dataset, val_dataset = torch.utils.data.random_split(dataset, [train_size, val_size])
+
+# Store the indices of the validation dataset
+val_indices = val_dataset.indices
+train_indices = train_dataset.indices
+
+train_loader = DataLoader(train_dataset, batch_size=1, shuffle=False)
+val_loader = DataLoader(val_dataset, batch_size=1, shuffle=False)
+
+
+# Tracking outputs for validation samples
+val_outputs = {}
+train_outputs = {}
+
+# Initialize model, loss, and optimizer
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+
+###########################################
+# THIS IS WHERE YOU CHOOSE A MODEL TO TEST 
+###########################################
+
+import importlib
+
+# Assuming sys.argv[1] is the module name
+module_name = sys.argv[1]  # Example: "my_model"
+class_name = "AgePredictionCNN"  # The class you want to import
+
+try:
+    # Dynamically import the module
+    module = importlib.import_module(module_name)
+    
+    # Dynamically get the class
+    AgePredictionCNN = getattr(module, class_name)
+    
+    print(f"Successfully imported {class_name} from {module_name}.")
+
+except ImportError:
+    print(f"Module {module_name} could not be imported.")
+except AttributeError:
+    print(f"{class_name} does not exist in {module_name}.")
+
+##############################
+# MODEL IMPORTED DYNAMICALLY
+##############################
+
+model = AgePredictionCNN(features_list[0].shape).to(device)
+criterion = nn.L1Loss()  # MAE Loss
+optimizer = optim.Adam(model.parameters(), lr=0.001)
+best_loss = np.inf  # Initialize the best loss to infinity
+start_epoch = 0
+
+load_saved = sys.argv[2] # "last, "best"
+if load_saved != "none":
+    # Load the checkpoint
+    with open(f"model_dumps/{sys.argv[1]}_{load_saved}_model_with_metadata_1kroi.pkl", "rb") as f:
+        checkpoint = pickle.load(f)
+
+    # Restore model and optimizer state
+    model.load_state_dict(checkpoint["model_state"])
+    optimizer.load_state_dict(checkpoint["optimizer_state"])
+
+    # Restore RNG states
+    torch.set_rng_state(checkpoint["t_rng_st"])
+    np.random.set_state(checkpoint["n_rng_st"])
+    if torch.cuda.is_available() and checkpoint["cuda_rng_st"] is not None:
+        torch.cuda.set_rng_state_all(checkpoint["cuda_rng_st"])
+
+    # Retrieve metadata
+    start_epoch = checkpoint["epoch"] + 1
+    loaded_loss = checkpoint["loss"]
+
+    print(f"Loaded model from epoch {start_epoch} with best validation loss: {loaded_loss:.4f}")
+
+predicted_ages = None
+# Training loop
+epochs = 200
+
+# Initialize lists to track loss
+filename = sys.argv[1] 
+csv_file = f"model_dumps/{filename}1kroi.csv"
+
+# Load existing epoch data if the file exists
+if os.path.exists(csv_file):
+    epoch_data = pd.read_csv(csv_file).to_dict(orient="records")
+    print(f"Loaded existing epoch data from {csv_file}.")
+else:
+    epoch_data = []
+    print("No existing epoch data found. Starting fresh.")
+
+
+# Plot loss vs. epoch and save the figure
+def update_loss_plot(epoch_data, filename):
+    df = pd.DataFrame(epoch_data)
+    df.to_csv(f"model_dumps/{filename}1kroi.csv", index=False)  # Save the data to CSV
+    
+    plt.figure(figsize=(8, 6))
+    plt.plot(df['epoch'], df['train_loss'], label="Train Loss", marker="o")
+    plt.plot(df['epoch'], df['val_loss'], label="Validation Loss", marker="o")
+    plt.xlabel("Epoch")
+    plt.ylabel("Loss")
+    plt.title("Loss vs. Epoch")
+    plt.legend()
+    plt.grid(True)
+    plt.savefig(f"model_dumps/{filename}1kroi.png")
+    plt.close()
+
+# Training loop
+for epoch in range(start_epoch, epochs):
+    model.train()
+    train_loss = 0.0
+    predicted_ages = []
+
+    for idx, (features, sex, age) in enumerate(train_loader):
+        features = features.unsqueeze(1).to(device)  # Add channel dimension
+        sex = sex.to(device)
+        age = age.to(device)
+        optimizer.zero_grad()
+        outputs = model(features, sex)
+        train_outputs[train_indices[idx]] = outputs.item()
+
+        loss = criterion(outputs.squeeze(), age)
+        loss.backward()
+        optimizer.step()
+
+        train_loss += loss.item()
+
+    train_loss /= len(train_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Train Loss: {train_loss:.4f}")
+
+    # Validation loop
+    model.eval()
+    val_loss = 0.0
+    with torch.no_grad():
+        for idx, (features, sex, age) in enumerate(val_loader):
+            features = features.unsqueeze(1).to(device)
+            sex = sex.to(device)
+            age = age.to(device)
+
+            outputs = model(features, sex)
+            loss = criterion(outputs.squeeze(), age)
+            val_loss += loss.item()
+
+            # Save the predicted age for the current validation sample
+            val_outputs[val_indices[idx]] = outputs.item()
+
+    val_loss /= len(val_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Validation Loss: {val_loss:.4f}")
+
+    # Save the last model with metadata
+    print(f"Saving last model...")
+    checkpoint = {
+        "model_state": model.state_dict(),
+        "optimizer_state": optimizer.state_dict(),
+        "epoch": epoch,
+        "loss": val_loss,
+        "t_rng_st": torch.get_rng_state(),
+        "n_rng_st": np.random.get_state(),
+        "cuda_rng_st": torch.cuda.get_rng_state_all() if torch.cuda.is_available() else None
+    }
+    with open(f"model_dumps/{sys.argv[1]}_last_model_with_metadata.pkl", "wb") as f:
+        pickle.dump(checkpoint, f)
+    print(f"Last model saved...")
+
+    # Check if validation loss improved
+    if val_loss < best_loss:
+        best_loss = val_loss
+        print(f"Validation loss improved to {best_loss:.4f}. Saving model...")
+        with open(f"model_dumps/{sys.argv[1]}_best_model_with_metadata1kroi.pkl", "wb") as f:
+            pickle.dump(checkpoint, f)
+
+    # Save predictions and create DataFrames (same as before)
+    # ...
+
+    # Update epoch data and save the loss plot
+    epoch_data.append({
+        "epoch": epoch + 1,
+        "train_loss": train_loss,
+        "val_loss": val_loss
+    })
+    update_loss_plot(epoch_data, sys.argv[1])
+
+    max_index = max(train_outputs.keys())
+    # Create a DataFrame with NaN for all indices initially
+    df_trn = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
+    # Assign the values to their respective indices
+    for index, value in train_outputs.items():
+        df_trn.loc[index, "Predicted_Age"] = value
+    # print (df_trn)
+
+    df2 = df.copy()
+    # df2['Predicted_Age'] = df_trn['Predicted_Age']
+    # train_df = df2.loc[train_outputs.keys()]
+    # print (train_df)
+    # train_df.to_csv(f"model_dumps/{sys.argv[1]}_predicted_ages_train1kroi.csv")
+
+    # max_index = max(val_outputs.keys())
+    # # Create a DataFrame with NaN for all indices initially
+    # df_pred = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
+    # # Assign the values to their respective indices
+    # for index, value in val_outputs.items():
+    #     df_pred.loc[index, "Predicted_Age"] = value
+    # # print (df_pred)
+
+    # df1 = df.copy()
+    # df1['Predicted_Age'] = df_pred['Predicted_Age']
+    # test_df = df1.loc[val_outputs.keys()]
+    # # print (test_df)
+    # test_df.to_csv(f"model_dumps/{sys.argv[1]}_predicted_ages_val1kroi.csv")
+
+
+    # Check that the predictions have been added to the DataFrame
+    # Plot Age vs. Predicted Age
+    # plt.figure(figsize=(8, 6))
+    # plt.scatter(train_df['Age'], train_df['Predicted_Age'], color='blue', label='Predicted vs Actual')
+    # # plt.plot(test_df['Age'], test_df['Age'], color='red', linestyle='--', label='Perfect Prediction')  # Optional: Line of perfect prediction
+    # plt.xlabel('Age')
+    # plt.ylabel('Predicted Age')
+    # plt.title('Age vs Predicted Age')
+    # plt.legend()
+    # plt.grid(True)
+    # plt.show()
\ No newline at end of file
diff --git a/vit_feature_cnn_main_mix.py b/vit_feature_cnn_main_mix.py
new file mode 100644
index 0000000..3ce1685
--- /dev/null
+++ b/vit_feature_cnn_main_mix.py
@@ -0,0 +1,1042 @@
+import os
+import pandas as pd
+import nibabel as nib
+import numpy as np
+import torch
+import torch.nn as nn
+import torch.nn.init as init
+import torch.optim as optim
+from tqdm import tqdm
+from nibabel.orientations import io_orientation, axcodes2ornt, ornt_transform, apply_orientation
+from torchvision import transforms
+from transformers import ViTFeatureExtractor, ViTModel
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler
+import pickle
+import matplotlib.pyplot as plt
+import sys
+import SimpleITK as sitk
+from scipy.ndimage import zoom
+
+def set_random_seed(seed=69420):
+    np.random.seed(seed)
+    torch.manual_seed(seed)
+    torch.cuda.manual_seed(seed)
+    torch.cuda.manual_seed_all(seed)  # For multi-GPU setups
+    torch.backends.cudnn.deterministic = True
+    torch.backends.cudnn.benchmark = False
+
+universal_seed = 69420
+
+set_random_seed(universal_seed)
+
+# Function to resample the volume to 160 slices
+def resample_volume_to_fixed_slices(data, affine, target_slices=160):
+    # Convert Numpy array and affine to SimpleITK image
+    sitk_img = sitk.GetImageFromArray(data)
+    sitk_img.SetSpacing([affine[0, 0], affine[1, 1], affine[2, 2]])
+
+    original_size = sitk_img.GetSize()  # (width, height, depth)
+    original_spacing = sitk_img.GetSpacing()  # (spacing_x, spacing_y, spacing_z)
+
+    # Calculate new spacing to achieve the target number of slices
+    new_spacing = list(original_spacing)
+    new_spacing[2] = (original_spacing[2] * original_size[2]) / target_slices
+
+    # Define new size
+    new_size = [original_size[0], original_size[1], target_slices]
+
+    # Resample the image
+    resampler = sitk.ResampleImageFilter()
+    resampler.SetOutputSpacing(new_spacing)
+    resampler.SetSize(new_size)
+    resampler.SetInterpolator(sitk.sitkLinear)
+    resampled_img = resampler.Execute(sitk_img)
+
+    return sitk.GetArrayFromImage(resampled_img)  # Return the resampled image as a numpy array
+
+
+def resample_nifti(img_data, target_slices = 160):
+    # Determine the current number of slices along the z-axis (3rd dimension)
+    current_slices = img_data.shape[0]
+    # Calculate the zoom factor for resampling (only along the z-axis)
+    zoom_factor = target_slices / current_slices
+    # Resample the image data along the z-axis
+    resampled_data = zoom(img_data, (zoom_factor, 1, 1), order=3)  # order=3 for cubic interpolation
+    # Ensure that the resampled data has the target number of slices
+    # print (resampled_data.shape)
+    # resampled_data = resampled_data[:target_slices,:,:]
+    # print (resampled_data.shape)
+    return resampled_data
+
+
+# Check if GPU is available
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+# Load the CSV file into a pandas DataFrame
+csv_path = "adni_storage/adni_brainrotnet_metadata.csv"
+df_adni = pd.read_csv(csv_path)
+# df = df.sample(n=1000, random_state=69420)
+# Add a new column 'filepath' with the constructed file paths
+df_adni['filepath'] = df_adni.apply(
+    lambda row: f"adni_storage/ADNI_nii_gz_bias_corrected/I{row['ImageID'][4:]}_{row['SubjectID']}.stripped.N4.nii.gz",
+    axis=1
+)
+df_adni = df_adni.sort_values(by='Age', ascending=True).reset_index(drop=True).head(500)
+# df_adni=df_adni.sample(n=400)
+
+# Load independent dataset metadata
+metadata_path = "ixi_storage/ixi_brainrotnet_metadata.csv"
+df_ixi = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_ixi['filepath'] = df_ixi.apply(
+    lambda row: f"ixi_storage/IXI_bias_corrected/{row['ImageID']}.stripped.N4.nii.gz",
+    axis=1
+)
+
+# metadata_path = "abide_storage/abide_brainrotnet_metadata.csv"
+# df_abide = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_abide['filepath'] = df_abide.apply(
+#     lambda row: f"abide_storage/ABIDEII_bias_corrected/{row['ImageID'][7:]}.stripped.N4.nii.gz",
+#     axis=1
+# )
+# df_abide = df_abide.sort_values(by='Age', ascending=False).reset_index(drop=True).head(500)
+# df_abide=df_abide.sample(n=400)
+
+metadata_path = "dlbs_storage/dlbs_brainrotnet_metadata.csv"
+df_dlbs = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_dlbs['filepath'] = df_dlbs.apply(
+    lambda row: f"dlbs_storage/DLBS_bias_corrected/{row['ImageID'][4:]}.stripped.N4.nii.gz",
+    axis=1
+)
+
+metadata_path = "camcan_storage/camcan_brainrotnet_metadata.csv"
+df_camcan = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_camcan['filepath'] = df_camcan.apply(
+    lambda row: f"camcan_storage/CamCAN_nii_gz_bias_corrected/{row['ImageID']}.stripped.N4.nii.gz",
+    axis=1
+)
+
+# metadata_path = "cobre_storage/cobre_brainrotnet_metadata.csv"
+# df_cobre = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_cobre['filepath'] = df_cobre.apply(
+#     lambda row: f"cobre_storage/COBRE_bias_corrected/{row['ImageID'][5:]}.stripped.N4.nii.gz",
+#     axis=1
+# )
+
+metadata_path = "fcon1000_storage/fcon1000_brainrotnet_metadata.csv"
+df_fcon = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_fcon['filepath'] = df_fcon.apply(
+    lambda row: f"fcon1000_storage/fcon1000_bias_corrected/{row['ImageID'][8:]}.stripped.N4.nii.gz",
+    axis=1
+)
+df_fcon = df_fcon.dropna()
+# df_fcon = df_fcon.sort_values(by='Age', ascending=False).reset_index(drop=True).head(300)
+# df_fcon = df_fcon.sample(n=300)
+
+# metadata_path = "sald_storage/sald_brainrotnet_metadata.csv"
+# df_sald = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_sald['filepath'] = df_sald.apply(
+#     lambda row: f"sald_storage/SALD_bias_corrected/sub-{row['ImageID'][4:]}.stripped.N4.nii.gz",
+#     axis=1
+# )
+# df_sald = df_sald.sort_values(by='Age', ascending=False).reset_index(drop=True).head(300)
+# # df_sald = df_sald.sample(n=300)
+
+# metadata_path = "corr_storage/corr_brainrotnet_metadata.csv"
+# df_corr = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_corr['filepath'] = df_corr.apply(
+#     lambda row: f"corr_storage/CORR_bias_corrected/{row['ImageID'][5:]}.stripped.N4.nii.gz",
+#     axis=1
+# )
+# df_corr = df_corr.sort_values(by='Age', ascending=True).reset_index(drop=True).head(300)
+# df_corr = df_corr.sample(n=200)
+
+
+# metadata_path = "oasis1_storage/oasis1_brainrotnet_metadata.csv"
+# df_oas1 = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_oas1['filepath'] = df_oas1.apply(
+#     lambda row: f"oasis1_storage/oasis_nii_gz_bias_corrected/{row['ImageID']}.stripped.N4.nii.gz",
+#     axis=1
+# )
+# df_oas1 = df_oas1.sort_values(by='Age', ascending=False).reset_index(drop=True).head(300)
+
+
+df = pd.concat ([
+                 df_adni[['ImageID', 'Sex', 'Age']], 
+                #  df_ixi[['ImageID', 'Sex', 'Age']], 
+                #  df_abide[['ImageID', 'Sex', 'Age']],
+                 df_dlbs[['ImageID', 'Sex', 'Age']],
+                #  df_cobre[['ImageID', 'Sex', 'Age']],
+                 df_fcon[['ImageID', 'Sex', 'Age']],
+                #  df_sald[['ImageID', 'Sex', 'Age']],
+                #  df_corr[['ImageID', 'Sex', 'Age']], 
+                #  df_oas1[['ImageID', 'Sex', 'Age']],
+                df_camcan[['ImageID', 'Sex', 'Age']]
+                 ], ignore_index=True)
+print (df)
+
+# Load pre-trained ViT model
+feature_extractor = ViTFeatureExtractor.from_pretrained("google/vit-base-patch16-224")
+model = ViTModel.from_pretrained("google/vit-base-patch16-224")
+model.to(device)  # Move the model to the GPU (if available)
+model.eval()
+
+# Update image transform for grayscale images to match ViT input requirements
+transform = transforms.Compose([
+    transforms.ToPILImage(),
+    transforms.Resize((224, 224)),
+    transforms.Lambda(lambda img: img.convert("RGB")),  # Convert to RGB directly
+    transforms.ToTensor(),
+    transforms.Normalize(mean=feature_extractor.image_mean, std=feature_extractor.image_std),
+])
+
+roi = 160
+
+
+# Directory to save processed images and features
+os.makedirs("adni_storage/ADNI_features/", exist_ok=True)
+
+# To store features and labels
+features_list = []
+labels_list = []
+
+# Process each row in the DataFrame
+for _, row in tqdm(df_adni.iterrows(), total=len(df_adni), desc="Processing images"):
+    filepath = row['filepath']
+    image_title = f"{row['ImageID'][4:]}_{row['SubjectID']}"
+
+    # Check if the feature file already exists
+    feature_file_path = f"adni_storage/ADNI_features/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        
+        # from PIL import Image
+        # # Normalize the array to 0-255 for grayscale image
+        # data_normalized = ((features - np.min(features)) / (np.max(features) - np.min(features)) * 255).astype(np.uint8)
+        # data_normalized = np.repeat(data_normalized, 4, axis=0)
+        # # Create an image from the array
+        # img = Image.fromarray(np.transpose(data_normalized), mode='L')  # 'L' mode for grayscale
+        # # Save the image
+        # img.save(f"adni_storage/ADNI_features/featuremaps/{image_title}_fm.png")
+
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        # print ("hiii")
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_volume_to_fixed_slices(data, affine, target_slices=160)
+
+                # Extract features for all sagittal slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+
+                    # Transform slice for ViT input
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)  # Add batch dimension and move to GPU
+
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        outputs = model(slice_tensor)
+                        slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()  # Move output back to CPU
+                        features.append(slice_features)
+
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Target is 'Age'
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+# for _, row in tqdm(df_ixi.iterrows(), total=len(df_ixi), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID']}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"ixi_storage/IXI_features/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         outputs = model(slice_tensor)
+#                         slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+# for _, row in tqdm(df_abide.iterrows(), total=len(df_abide), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][7:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"abide_storage/ABIDEII_features/{image_title}_features.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         outputs = model(slice_tensor)
+#                         slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+for _, row in tqdm(df_dlbs.iterrows(), total=len(df_dlbs), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID'][4:]}"
+        # Check if the feature file already exists
+    feature_file_path = f"dlbs_storage/DLBS_features/{image_title}_features.npy"
+    # feature_file_path = f"ixi_storage/IXI_features/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        outputs = model(slice_tensor)
+                        slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+
+
+
+# for _, row in tqdm(df_cobre.iterrows(), total=len(df_cobre), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][5:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"cobre_storage/COBRE_features/{image_title}_features.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         outputs = model(slice_tensor)
+#                         slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+
+for _, row in tqdm(df_fcon.iterrows(), total=len(df_fcon), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID'][5:]}"
+        # Check if the feature file already exists
+    feature_file_path = f"fcon1000_storage/fcon1000_features/{image_title}_features.npy"
+    # feature_file_path = f"ixi_storage/IXI_features/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        outputs = model(slice_tensor)
+                        slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+# for _, row in tqdm(df_sald.iterrows(), total=len(df_sald), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][4:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"sald_storage/SALD_features/{image_title}_features.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         outputs = model(slice_tensor)
+#                         slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+# for _, row in tqdm(df_corr.iterrows(), total=len(df_corr), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][5:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"corr_storage/CORR_features/{image_title}_features.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         outputs = model(slice_tensor)
+#                         slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+
+# for _, row in tqdm(df_oas1.iterrows(), total=len(df_oas1), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID']}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"oasis1_storage/oasis1_features/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         outputs = model(slice_tensor)
+#                         slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+for _, row in tqdm(df_camcan.iterrows(), total=len(df_camcan), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID']}"
+        # Check if the feature file already exists
+    feature_file_path = f"camcan_storage/CamCAN_features/{image_title}_features.npy"
+    # feature_file_path = f"ixi_storage/IXI_features/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        outputs = model(slice_tensor)
+                        slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, Dataset
+from dataset_cls import ADNIDataset
+
+# Prepare dataset and dataloaders
+sex_encoded = df['Sex'].apply(lambda x: 0 if x == 'M' else 1).tolist()
+age_list = df['Age'].tolist()
+
+batch_size = 1
+
+# print (features_list)
+print (features_list[0].shape)
+
+# Create Dataset and DataLoader
+dataset = ADNIDataset(features_list, sex_encoded, age_list)
+train_size = int(0.8 * len(dataset))
+val_size = len(dataset) - train_size
+train_dataset, val_dataset = torch.utils.data.random_split(dataset, [train_size, val_size])
+
+# Store the indices of the validation dataset
+val_indices = val_dataset.indices
+train_indices = train_dataset.indices
+
+train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=False)
+val_loader = DataLoader(val_dataset, batch_size=batch_size, shuffle=False)
+
+
+# Tracking outputs for validation samples
+val_outputs = {}
+train_outputs = {}
+
+# Initialize model, loss, and optimizer
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+
+###########################################
+# THIS IS WHERE YOU CHOOSE A MODEL TO TEST 
+###########################################
+
+import importlib
+
+# Assuming sys.argv[1] is the module name
+module_name = sys.argv[1]  # Example: "my_model"
+class_name = "AgePredictionCNN"  # The class you want to import
+
+try:
+    # Dynamically import the module
+    module = importlib.import_module(module_name)
+    
+    # Dynamically get the class
+    AgePredictionCNN = getattr(module, class_name)
+    
+    print(f"Successfully imported {class_name} from {module_name}.")
+
+except ImportError:
+    print(f"Module {module_name} could not be imported.")
+except AttributeError:
+    print(f"{class_name} does not exist in {module_name}.")
+
+##############################
+# MODEL IMPORTED DYNAMICALLY
+##############################
+
+print (features_list[0].shape)
+model = AgePredictionCNN((1, features_list[0].shape[0], features_list[0].shape[1])).to(device)
+criterion = nn.MSELoss()  # MAE Loss
+eval_crit = nn.L1Loss()
+optimizer = optim.Adam(model.parameters(), lr=0.001)
+best_loss = np.inf  # Initialize the best loss to infinity
+start_epoch = 0
+
+
+load_saved = sys.argv[2] # "last, "best"
+if load_saved != "none":
+    with open(f"model_dumps/mix/{sys.argv[1]}_best_model_with_metadata.pkl", "rb") as f:
+        checkpoint = pickle.load(f)
+    best_loss = checkpoint["loss"]
+
+    # Load the checkpoint
+    with open(f"model_dumps/mix/{sys.argv[1]}_{load_saved}_model_with_metadata.pkl", "rb") as f:
+        checkpoint = pickle.load(f)
+
+    # Restore model and optimizer state
+    model.load_state_dict(checkpoint["model_state"])
+    optimizer.load_state_dict(checkpoint["optimizer_state"])
+
+    # Restore RNG states
+    torch.set_rng_state(checkpoint["t_rng_st"])
+    np.random.set_state(checkpoint["n_rng_st"])
+    if torch.cuda.is_available() and checkpoint["cuda_rng_st"] is not None:
+        torch.cuda.set_rng_state_all(checkpoint["cuda_rng_st"])
+
+    # Retrieve metadata
+    start_epoch = checkpoint["epoch"] + 1
+    loaded_loss = checkpoint["loss"]
+
+    print(f"Loaded model from epoch {start_epoch} with validation loss {loaded_loss:.4f}, best loss {best_loss:.4f}")
+
+predicted_ages = None
+# Training loop
+epochs = int(sys.argv[3])
+
+# Initialize lists to track loss
+filename = sys.argv[1] 
+csv_file = f"model_dumps/mix/{filename}.csv"
+
+# Load existing epoch data if the file exists
+if os.path.exists(csv_file):
+    epoch_data = pd.read_csv(csv_file).to_dict(orient="records")
+    print(f"Loaded existing epoch data from {csv_file}.")
+else:
+    epoch_data = []
+    print("No existing epoch data found. Starting fresh.")
+
+
+# Plot loss vs. epoch and save the figure
+def update_loss_plot(epoch_data, filename):
+    df = pd.DataFrame(epoch_data)
+    df.to_csv(f"model_dumps/mix/{filename}.csv", index=False)  # Save the data to CSV
+    
+    plt.figure(figsize=(8, 6))
+    plt.plot(df['epoch'], df['train_loss'], label="Train Loss", marker="o")
+    plt.plot(df['epoch'], df['val_loss'], label="Validation Loss", marker="o")
+    plt.xlabel("Epoch")
+    plt.ylabel("Loss")
+    plt.title("Loss vs. Epoch")
+    plt.legend()
+    plt.grid(True)
+    plt.savefig(f"model_dumps/mix/{filename}.png")
+    plt.close()
+
+# Training loop
+for epoch in range(start_epoch, epochs):
+    model.train()
+    train_loss = 0.0
+    predicted_ages = []
+
+    for idx, (features, sex, age) in enumerate(train_loader):
+        features = features.unsqueeze(1).to(device)  # Add channel dimension
+        sex = sex.to(device)
+        age = age.to(device)
+        optimizer.zero_grad()
+        outputs = model(features, sex)
+        # Store the output for each sample in the batch
+        for i in range(outputs.size(0)):
+            train_outputs[train_indices[idx * batch_size + i]] = outputs[i].item()
+
+        loss = criterion(outputs.squeeze(), age)
+        loss.backward()
+        optimizer.step()
+
+        train_loss += loss.item()
+
+    train_loss /= len(train_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Train Loss: {train_loss:.4f}")
+
+    # Validation loop
+    model.eval()
+    val_loss = 0.0
+    with torch.no_grad():
+        for idx, (features, sex, age) in enumerate(val_loader):
+            features = features.unsqueeze(1).to(device)
+            sex = sex.to(device)
+            age = age.to(device)
+
+            outputs = model(features, sex)
+            loss = eval_crit(outputs.squeeze(), age)
+            val_loss += loss.item()
+
+            # Save the predicted age for the current validation sample
+            for i in range(outputs.size(0)):
+                val_outputs[val_indices[idx * batch_size + i]] = outputs[i].item()
+            # val_outputs[val_indices[idx]] = outputs.item()
+
+    val_loss /= len(val_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Validation Loss: {val_loss:.4f}")
+
+    # Save the last model with metadata
+    print(f"Saving last model...")
+    checkpoint = {
+        "model_state": model.state_dict(),
+        "optimizer_state": optimizer.state_dict(),
+        "epoch": epoch,
+        "loss": val_loss,
+        "t_rng_st": torch.get_rng_state(),
+        "n_rng_st": np.random.get_state(),
+        "cuda_rng_st": torch.cuda.get_rng_state_all() if torch.cuda.is_available() else None
+    }
+    with open(f"model_dumps/mix/{sys.argv[1]}_last_model_with_metadata.pkl", "wb") as f:
+        pickle.dump(checkpoint, f)
+    print(f"Last model saved...")
+
+    # Check if validation loss improved
+    if val_loss < best_loss:
+        best_loss = val_loss
+        print(f"Validation loss improved to {best_loss:.4f}. Saving model...")
+        with open(f"model_dumps/mix/{sys.argv[1]}_best_model_with_metadata.pkl", "wb") as f:
+            pickle.dump(checkpoint, f)
+
+    # Save predictions and create DataFrames (same as before)
+    # ...
+
+    # Update epoch data and save the loss plot
+    epoch_data.append({
+        "epoch": epoch + 1,
+        "train_loss": train_loss,
+        "val_loss": val_loss
+    })
+    update_loss_plot(epoch_data, sys.argv[1])
+
+    max_index = max(train_outputs.keys())
+    # Create a DataFrame with NaN for all indices initially
+    df_trn = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
+    # Assign the values to their respective indices
+    for index, value in train_outputs.items():
+        df_trn.loc[index, "Predicted_Age"] = value
+    # print (df_trn)
+
+    df2 = df.copy()
+    df2['Predicted_Age'] = df_trn['Predicted_Age']
+    train_df = df2.loc[train_outputs.keys()]
+    # print (train_df)
+    train_df.to_csv(f"model_dumps/mix/{sys.argv[1]}_predicted_ages_train.csv")
+
+    max_index = max(val_outputs.keys())
+    # Create a DataFrame with NaN for all indices initially
+    df_pred = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
+    # Assign the values to their respective indices
+    for index, value in val_outputs.items():
+        df_pred.loc[index, "Predicted_Age"] = value
+    # print (df_pred)
+
+    df1 = df.copy()
+    df1['Predicted_Age'] = df_pred['Predicted_Age']
+    test_df = df1.loc[val_outputs.keys()]
+    print (test_df)
+    test_df.to_csv(f"model_dumps/mix/{sys.argv[1]}_predicted_ages_val.csv")
+
+
+    # Map unique first 4 characters of ImageID to color codes
+    unique_groups = test_df['ImageID'].str[:3].unique()
+    group_to_color = {group: i for i, group in enumerate(unique_groups)}
+
+    # Assign colors based on the mapping
+    cmap = plt.get_cmap('tab10')  # Change colormap as desired
+    colors = [cmap(group_to_color[group]) for group in test_df['ImageID'].str[:3]]
+
+    # Check that the predictions have been added to the DataFrame
+    # Plot Age vs. Predicted Age
+    plt.figure(figsize=(8, 6))
+    plt.scatter(test_df['Age'], test_df['Predicted_Age'], color=colors, label='Predicted vs Actual')
+    # plt.plot(test_df['Age'], test_df['Age'], color='red', linestyle='--', label='Perfect Prediction')  # Optional: Line of perfect prediction
+    # Add legend for colors based on ImageID groups
+    handles = [plt.Line2D([0], [0], marker='o', color=cmap(i), linestyle='', markersize=10) 
+            for i, group in enumerate(unique_groups)]
+    plt.legend(handles, unique_groups, title="ImageID Groups")
+    plt.xlabel('Age')
+    plt.ylabel('Predicted Age')
+    plt.title('Age vs Predicted Age')
+    plt.grid(True)
+    plt.savefig(f"model_dumps/mix/plots/vit_cnn_{sys.argv[1]}_epoch{epoch}.png")
diff --git a/vit_feature_cnn_main_mix_roi.py b/vit_feature_cnn_main_mix_roi.py
new file mode 100644
index 0000000..efc912e
--- /dev/null
+++ b/vit_feature_cnn_main_mix_roi.py
@@ -0,0 +1,1012 @@
+import os
+import pandas as pd
+import nibabel as nib
+import numpy as np
+import torch
+import torch.nn as nn
+import torch.nn.init as init
+import torch.optim as optim
+from tqdm import tqdm
+from nibabel.orientations import io_orientation, axcodes2ornt, ornt_transform, apply_orientation
+from torchvision import transforms
+from transformers import ViTFeatureExtractor, ViTModel
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler
+import pickle
+import matplotlib.pyplot as plt
+import sys
+import SimpleITK as sitk
+from scipy.ndimage import zoom
+
+def set_random_seed(seed=69420):
+    np.random.seed(seed)
+    torch.manual_seed(seed)
+    torch.cuda.manual_seed(seed)
+    torch.cuda.manual_seed_all(seed)  # For multi-GPU setups
+    torch.backends.cudnn.deterministic = True
+    torch.backends.cudnn.benchmark = False
+
+universal_seed = 69420
+
+set_random_seed(universal_seed)
+
+# Function to resample the volume to 160 slices
+def resample_volume_to_fixed_slices(data, affine, target_slices=160):
+    # Convert Numpy array and affine to SimpleITK image
+    sitk_img = sitk.GetImageFromArray(data)
+    sitk_img.SetSpacing([affine[0, 0], affine[1, 1], affine[2, 2]])
+
+    original_size = sitk_img.GetSize()  # (width, height, depth)
+    original_spacing = sitk_img.GetSpacing()  # (spacing_x, spacing_y, spacing_z)
+
+    # Calculate new spacing to achieve the target number of slices
+    new_spacing = list(original_spacing)
+    new_spacing[2] = (original_spacing[2] * original_size[2]) / target_slices
+
+    # Define new size
+    new_size = [original_size[0], original_size[1], target_slices]
+
+    # Resample the image
+    resampler = sitk.ResampleImageFilter()
+    resampler.SetOutputSpacing(new_spacing)
+    resampler.SetSize(new_size)
+    resampler.SetInterpolator(sitk.sitkLinear)
+    resampled_img = resampler.Execute(sitk_img)
+
+    return sitk.GetArrayFromImage(resampled_img)  # Return the resampled image as a numpy array
+
+
+def resample_nifti(img_data, target_slices = 160):
+    # Determine the current number of slices along the z-axis (3rd dimension)
+    current_slices = img_data.shape[0]
+    # Calculate the zoom factor for resampling (only along the z-axis)
+    zoom_factor = target_slices / current_slices
+    # Resample the image data along the z-axis
+    resampled_data = zoom(img_data, (zoom_factor, 1, 1), order=3)  # order=3 for cubic interpolation
+    # Ensure that the resampled data has the target number of slices
+    # print (resampled_data.shape)
+    # resampled_data = resampled_data[:target_slices,:,:]
+    # print (resampled_data.shape)
+    return resampled_data
+
+def calculate_bounding_box_from_volume(volume):
+    # Find indices of non-zero values
+    non_zero_indices = np.argwhere(volume > 0)
+
+    # Calculate min and max indices along each dimension
+    min_indices = np.min(non_zero_indices, axis=0)
+    max_indices = np.max(non_zero_indices, axis=0)
+
+    # Convert indices to integers
+    min_indices = min_indices.astype(int)
+    max_indices = max_indices.astype(int)
+
+    return min_indices, max_indices
+
+
+def crop_brain_volumes(brain_data):
+    
+
+        # Calculate bounding box from the brain volume
+    min_indices, max_indices = calculate_bounding_box_from_volume(brain_data)
+
+        # Crop the volume
+    cropped_brain = brain_data[min_indices[0]:max_indices[0] + 1,
+                                   min_indices[1]:max_indices[1] + 1,
+                                   min_indices[2]:max_indices[2] + 1]
+    return cropped_brain
+
+
+# Check if GPU is available
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+# Load the CSV file into a pandas DataFrame
+csv_path = "adni_storage/adni_brainrotnet_metadata.csv"
+df_adni = pd.read_csv(csv_path)
+# df = df.sample(n=1000, random_state=69420)
+# Add a new column 'filepath' with the constructed file paths
+df_adni['filepath'] = df_adni.apply(
+    lambda row: f"adni_storage/ADNI_nii_gz_bias_corrected/I{row['ImageID'][4:]}_{row['SubjectID']}.stripped.N4.nii.gz",
+    axis=1
+)
+df_adni = df_adni.sort_values(by='Age', ascending=True).reset_index(drop=True).head(500)
+# df_adni=df_adni.sample(n=400)
+
+# Load independent dataset metadata
+metadata_path = "ixi_storage/ixi_brainrotnet_metadata.csv"
+df_ixi = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_ixi['filepath'] = df_ixi.apply(
+    lambda row: f"ixi_storage/IXI_bias_corrected/{row['ImageID']}.stripped.N4.nii.gz",
+    axis=1
+)
+
+metadata_path = "abide_storage/abide_brainrotnet_metadata.csv"
+df_abide = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_abide['filepath'] = df_abide.apply(
+    lambda row: f"abide_storage/ABIDEII_bias_corrected/{row['ImageID'][7:]}.stripped.N4.nii.gz",
+    axis=1
+)
+df_abide = df_abide.sort_values(by='Age', ascending=False).reset_index(drop=True).head(500)
+# df_abide=df_abide.sample(n=400)
+
+metadata_path = "dlbs_storage/dlbs_brainrotnet_metadata.csv"
+df_dlbs = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_dlbs['filepath'] = df_dlbs.apply(
+    lambda row: f"dlbs_storage/DLBS_bias_corrected/{row['ImageID'][4:]}.stripped.N4.nii.gz",
+    axis=1
+)
+
+metadata_path = "cobre_storage/cobre_brainrotnet_metadata.csv"
+df_cobre = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_cobre['filepath'] = df_cobre.apply(
+    lambda row: f"cobre_storage/COBRE_bias_corrected/{row['ImageID'][5:]}.stripped.N4.nii.gz",
+    axis=1
+)
+
+metadata_path = "fcon1000_storage/fcon1000_brainrotnet_metadata.csv"
+df_fcon = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_fcon['filepath'] = df_fcon.apply(
+    lambda row: f"fcon1000_storage/fcon1000_bias_corrected/{row['ImageID'][8:]}.stripped.N4.nii.gz",
+    axis=1
+)
+df_fcon = df_fcon.dropna()
+# df_fcon = df_fcon.sort_values(by='Age', ascending=False).reset_index(drop=True).head(300)
+# df_fcon = df_fcon.sample(n=300)
+
+metadata_path = "sald_storage/sald_brainrotnet_metadata.csv"
+df_sald = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_sald['filepath'] = df_sald.apply(
+    lambda row: f"sald_storage/SALD_bias_corrected/sub-{row['ImageID'][4:]}.stripped.N4.nii.gz",
+    axis=1
+)
+# df_sald = df_sald.sort_values(by='Age', ascending=False).reset_index(drop=True).head(300)
+# # df_sald = df_sald.sample(n=300)
+
+metadata_path = "corr_storage/corr_brainrotnet_metadata.csv"
+df_corr = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_corr['filepath'] = df_corr.apply(
+    lambda row: f"corr_storage/CORR_bias_corrected/{row['ImageID'][5:]}.stripped.N4.nii.gz",
+    axis=1
+)
+df_corr = df_corr.sort_values(by='Age', ascending=True).reset_index(drop=True).head(300)
+df_corr = df_corr.sample(n=200)
+
+
+metadata_path = "oasis1_storage/oasis1_brainrotnet_metadata.csv"
+df_oas1 = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_oas1['filepath'] = df_oas1.apply(
+    lambda row: f"oasis1_storage/oasis_nii_gz_bias_corrected/{row['ImageID']}.stripped.N4.nii.gz",
+    axis=1
+)
+df_oas1 = df_oas1.sort_values(by='Age', ascending=False).reset_index(drop=True).head(300)
+
+
+df = pd.concat ([
+                 df_adni[['ImageID', 'Sex', 'Age']], 
+                #  df_ixi[['ImageID', 'Sex', 'Age']], 
+                #  df_abide[['ImageID', 'Sex', 'Age']],
+                 df_dlbs[['ImageID', 'Sex', 'Age']],
+                #  df_cobre[['ImageID', 'Sex', 'Age']],
+                 df_fcon[['ImageID', 'Sex', 'Age']],
+                 df_sald[['ImageID', 'Sex', 'Age']],
+                #  df_corr[['ImageID', 'Sex', 'Age']], 
+                df_oas1[['ImageID', 'Sex', 'Age']],
+                 ], ignore_index=True)
+print (df)
+
+# Load pre-trained ViT model
+feature_extractor = ViTFeatureExtractor.from_pretrained("google/vit-base-patch16-224")
+model = ViTModel.from_pretrained("google/vit-base-patch16-224")
+model.to(device)  # Move the model to the GPU (if available)
+model.eval()
+
+# Update image transform for grayscale images to match ViT input requirements
+transform = transforms.Compose([
+    transforms.ToPILImage(),
+    transforms.Resize((224, 224)),
+    transforms.Lambda(lambda img: img.convert("RGB")),  # Convert to RGB directly
+    transforms.ToTensor(),
+    transforms.Normalize(mean=feature_extractor.image_mean, std=feature_extractor.image_std),
+])
+
+roi = 160
+
+
+# Directory to save processed images and features
+os.makedirs("adni_storage/ADNI_features/", exist_ok=True)
+
+# To store features and labels
+features_list = []
+labels_list = []
+
+# Process each row in the DataFrame
+for _, row in tqdm(df_adni.iterrows(), total=len(df_adni), desc="Processing images"):
+    filepath = row['filepath']
+    image_title = f"{row['ImageID'][4:]}_{row['SubjectID']}"
+
+    # Check if the feature file already exists
+    feature_file_path = f"adni_storage/ADNI_features/{image_title}_features_roi.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        
+        from PIL import Image
+        # Normalize the array to 0-255 for grayscale image
+        data_normalized = ((features - np.min(features)) / (np.max(features) - np.min(features)) * 255).astype(np.uint8)
+        data_normalized = np.repeat(data_normalized, 4, axis=0)
+        # Create an image from the array
+        img = Image.fromarray(np.transpose(data_normalized), mode='L')  # 'L' mode for grayscale
+        # Save the image
+        img.save(f"adni_storage/ADNI_features/featuremaps/{image_title}_fm.png")
+
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        # print ("hiii")
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Resample the volume to 160 slices (if required)
+                data = crop_brain_volumes(data)
+                data = resample_nifti(data, target_slices=160)
+
+                # Extract features for all sagittal slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+
+                    # Transform slice for ViT input
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)  # Add batch dimension and move to GPU
+
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        outputs = model(slice_tensor)
+                        slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()  # Move output back to CPU
+                        features.append(slice_features)
+
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Target is 'Age'
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+# for _, row in tqdm(df_ixi.iterrows(), total=len(df_ixi), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID']}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"ixi_storage/IXI_features/{image_title}_features_roi.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = crop_brain_volumes(data)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         outputs = model(slice_tensor)
+#                         slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+# for _, row in tqdm(df_abide.iterrows(), total=len(df_abide), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][7:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"abide_storage/ABIDEII_features/{image_title}_features_roi.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/{image_title}_features_roi.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = crop_brain_volumes(data)
+# data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         outputs = model(slice_tensor)
+#                         slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+for _, row in tqdm(df_dlbs.iterrows(), total=len(df_dlbs), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID'][4:]}"
+        # Check if the feature file already exists
+    feature_file_path = f"dlbs_storage/DLBS_features/{image_title}_features_roi.npy"
+    # feature_file_path = f"ixi_storage/IXI_features/{image_title}_features_roi.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Resample the volume to 160 slices (if required)
+                data = crop_brain_volumes(data)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        outputs = model(slice_tensor)
+                        slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+
+# for _, row in tqdm(df_cobre.iterrows(), total=len(df_cobre), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][5:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"cobre_storage/COBRE_features/{image_title}_features_roi.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/{image_title}_features_roi.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = crop_brain_volumes(data)
+# data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         outputs = model(slice_tensor)
+#                         slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+
+for _, row in tqdm(df_fcon.iterrows(), total=len(df_fcon), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID'][5:]}"
+        # Check if the feature file already exists
+    feature_file_path = f"fcon1000_storage/fcon1000_features/{image_title}_features_roi.npy"
+    # feature_file_path = f"ixi_storage/IXI_features/{image_title}_features_roi.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Resample the volume to 160 slices (if required)
+                data = crop_brain_volumes(data)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        outputs = model(slice_tensor)
+                        slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+for _, row in tqdm(df_sald.iterrows(), total=len(df_sald), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID'][4:]}"
+        # Check if the feature file already exists
+    feature_file_path = f"sald_storage/SALD_features/{image_title}_features_roi.npy"
+    # feature_file_path = f"ixi_storage/IXI_features/{image_title}_features_roi.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Resample the volume to 160 slices (if required)
+                data = crop_brain_volumes(data)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        outputs = model(slice_tensor)
+                        slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+# for _, row in tqdm(df_corr.iterrows(), total=len(df_corr), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][5:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"corr_storage/CORR_features/{image_title}_features_roi.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/{image_title}_features_roi.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = crop_brain_volumes(data)
+# data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         outputs = model(slice_tensor)
+#                         slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+
+for _, row in tqdm(df_oas1.iterrows(), total=len(df_oas1), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID']}"
+        # Check if the feature file already exists
+    feature_file_path = f"oasis1_storage/oasis1_features/{image_title}_features_roi.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Resample the volume to 160 slices (if required)
+                data = crop_brain_volumes(data)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        outputs = model(slice_tensor)
+                        slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, Dataset
+from dataset_cls import ADNIDataset
+
+# Prepare dataset and dataloaders
+sex_encoded = df['Sex'].apply(lambda x: 0 if x == 'M' else 1).tolist()
+age_list = df['Age'].tolist()
+
+batch_size = 1
+
+# print (features_list)
+print (features_list[0].shape)
+
+# Create Dataset and DataLoader
+dataset = ADNIDataset(features_list, sex_encoded, age_list)
+train_size = int(0.8 * len(dataset))
+val_size = len(dataset) - train_size
+train_dataset, val_dataset = torch.utils.data.random_split(dataset, [train_size, val_size])
+
+# Store the indices of the validation dataset
+val_indices = val_dataset.indices
+train_indices = train_dataset.indices
+
+train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=False)
+val_loader = DataLoader(val_dataset, batch_size=batch_size, shuffle=False)
+
+
+# Tracking outputs for validation samples
+val_outputs = {}
+train_outputs = {}
+
+# Initialize model, loss, and optimizer
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+
+###########################################
+# THIS IS WHERE YOU CHOOSE A MODEL TO TEST 
+###########################################
+
+import importlib
+
+# Assuming sys.argv[1] is the module name
+module_name = sys.argv[1]  # Example: "my_model"
+class_name = "AgePredictionCNN"  # The class you want to import
+
+try:
+    # Dynamically import the module
+    module = importlib.import_module(module_name)
+    
+    # Dynamically get the class
+    AgePredictionCNN = getattr(module, class_name)
+    
+    print(f"Successfully imported {class_name} from {module_name}.")
+
+except ImportError:
+    print(f"Module {module_name} could not be imported.")
+except AttributeError:
+    print(f"{class_name} does not exist in {module_name}.")
+
+##############################
+# MODEL IMPORTED DYNAMICALLY
+##############################
+
+print (features_list[0].shape)
+model = AgePredictionCNN((1, features_list[0].shape[0], features_list[0].shape[1])).to(device)
+criterion = nn.L1Loss()  # MAE Loss
+eval_crit = nn.L1Loss()
+optimizer = optim.Adam(model.parameters(), lr=0.001)
+best_loss = np.inf  # Initialize the best loss to infinity
+start_epoch = 0
+
+
+load_saved = sys.argv[2] # "last, "best"
+if load_saved != "none":
+    with open(f"model_dumps/mix/{sys.argv[1]}_best_model_with_metadata.pkl", "rb") as f:
+        checkpoint = pickle.load(f)
+    best_loss = checkpoint["loss"]
+
+    # Load the checkpoint
+    with open(f"model_dumps/mix/{sys.argv[1]}_{load_saved}_model_with_metadata.pkl", "rb") as f:
+        checkpoint = pickle.load(f)
+
+    # Restore model and optimizer state
+    model.load_state_dict(checkpoint["model_state"])
+    optimizer.load_state_dict(checkpoint["optimizer_state"])
+
+    # Restore RNG states
+    torch.set_rng_state(checkpoint["t_rng_st"])
+    np.random.set_state(checkpoint["n_rng_st"])
+    if torch.cuda.is_available() and checkpoint["cuda_rng_st"] is not None:
+        torch.cuda.set_rng_state_all(checkpoint["cuda_rng_st"])
+
+    # Retrieve metadata
+    start_epoch = checkpoint["epoch"] + 1
+    loaded_loss = checkpoint["loss"]
+
+    print(f"Loaded model from epoch {start_epoch} with validation loss {loaded_loss:.4f}, best loss {best_loss:.4f}")
+
+predicted_ages = None
+# Training loop
+epochs = int(sys.argv[3])
+
+# Initialize lists to track loss
+filename = sys.argv[1] 
+csv_file = f"model_dumps/mix/{filename}.csv"
+
+# Load existing epoch data if the file exists
+if os.path.exists(csv_file):
+    epoch_data = pd.read_csv(csv_file).to_dict(orient="records")
+    print(f"Loaded existing epoch data from {csv_file}.")
+else:
+    epoch_data = []
+    print("No existing epoch data found. Starting fresh.")
+
+
+# Plot loss vs. epoch and save the figure
+def update_loss_plot(epoch_data, filename):
+    df = pd.DataFrame(epoch_data)
+    df.to_csv(f"model_dumps/mix/{filename}.csv", index=False)  # Save the data to CSV
+    
+    plt.figure(figsize=(8, 6))
+    plt.plot(df['epoch'], df['train_loss'], label="Train Loss", marker="o")
+    plt.plot(df['epoch'], df['val_loss'], label="Validation Loss", marker="o")
+    plt.xlabel("Epoch")
+    plt.ylabel("Loss")
+    plt.title("Loss vs. Epoch")
+    plt.legend()
+    plt.grid(True)
+    plt.savefig(f"model_dumps/mix/{filename}.png")
+    plt.close()
+
+# Training loop
+for epoch in range(start_epoch, epochs):
+    model.train()
+    train_loss = 0.0
+    predicted_ages = []
+
+    for idx, (features, sex, age) in enumerate(train_loader):
+        features = features.unsqueeze(1).to(device)  # Add channel dimension
+        sex = sex.to(device)
+        age = age.to(device)
+        optimizer.zero_grad()
+        outputs = model(features, sex)
+        # Store the output for each sample in the batch
+        for i in range(outputs.size(0)):
+            train_outputs[train_indices[idx * batch_size + i]] = outputs[i].item()
+
+        loss = criterion(outputs.squeeze(), age)
+        loss.backward()
+        optimizer.step()
+
+        train_loss += loss.item()
+
+    train_loss /= len(train_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Train Loss: {train_loss:.4f}")
+
+    # Validation loop
+    model.eval()
+    val_loss = 0.0
+    with torch.no_grad():
+        for idx, (features, sex, age) in enumerate(val_loader):
+            features = features.unsqueeze(1).to(device)
+            sex = sex.to(device)
+            age = age.to(device)
+
+            outputs = model(features, sex)
+            loss = eval_crit(outputs.squeeze(), age)
+            val_loss += loss.item()
+
+            # Save the predicted age for the current validation sample
+            for i in range(outputs.size(0)):
+                val_outputs[val_indices[idx * batch_size + i]] = outputs[i].item()
+            # val_outputs[val_indices[idx]] = outputs.item()
+
+    val_loss /= len(val_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Validation Loss: {val_loss:.4f}")
+
+    # Save the last model with metadata
+    print(f"Saving last model...")
+    checkpoint = {
+        "model_state": model.state_dict(),
+        "optimizer_state": optimizer.state_dict(),
+        "epoch": epoch,
+        "loss": val_loss,
+        "t_rng_st": torch.get_rng_state(),
+        "n_rng_st": np.random.get_state(),
+        "cuda_rng_st": torch.cuda.get_rng_state_all() if torch.cuda.is_available() else None
+    }
+    with open(f"model_dumps/mix/{sys.argv[1]}_last_model_with_metadata.pkl", "wb") as f:
+        pickle.dump(checkpoint, f)
+    print(f"Last model saved...")
+
+    # Check if validation loss improved
+    if val_loss < best_loss:
+        best_loss = val_loss
+        print(f"Validation loss improved to {best_loss:.4f}. Saving model...")
+        with open(f"model_dumps/mix/{sys.argv[1]}_best_model_with_metadata.pkl", "wb") as f:
+            pickle.dump(checkpoint, f)
+
+    # Save predictions and create DataFrames (same as before)
+    # ...
+
+    # Update epoch data and save the loss plot
+    epoch_data.append({
+        "epoch": epoch + 1,
+        "train_loss": train_loss,
+        "val_loss": val_loss
+    })
+    update_loss_plot(epoch_data, sys.argv[1])
+
+    max_index = max(train_outputs.keys())
+    # Create a DataFrame with NaN for all indices initially
+    df_trn = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
+    # Assign the values to their respective indices
+    for index, value in train_outputs.items():
+        df_trn.loc[index, "Predicted_Age"] = value
+    # print (df_trn)
+
+    df2 = df.copy()
+    df2['Predicted_Age'] = df_trn['Predicted_Age']
+    train_df = df2.loc[train_outputs.keys()]
+    # print (train_df)
+    train_df.to_csv(f"model_dumps/mix/{sys.argv[1]}_predicted_ages_train.csv")
+
+    max_index = max(val_outputs.keys())
+    # Create a DataFrame with NaN for all indices initially
+    df_pred = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
+    # Assign the values to their respective indices
+    for index, value in val_outputs.items():
+        df_pred.loc[index, "Predicted_Age"] = value
+    # print (df_pred)
+
+    df1 = df.copy()
+    df1['Predicted_Age'] = df_pred['Predicted_Age']
+    test_df = df1.loc[val_outputs.keys()]
+    print (test_df)
+    test_df.to_csv(f"model_dumps/mix/{sys.argv[1]}_predicted_ages_val.csv")
+
+
+    # Map unique first 4 characters of ImageID to color codes
+    unique_groups = test_df['ImageID'].str[:3].unique()
+    group_to_color = {group: i for i, group in enumerate(unique_groups)}
+
+    # Assign colors based on the mapping
+    cmap = plt.get_cmap('tab10')  # Change colormap as desired
+    colors = [cmap(group_to_color[group]) for group in test_df['ImageID'].str[:3]]
+
+    # Check that the predictions have been added to the DataFrame
+    # Plot Age vs. Predicted Age
+    plt.figure(figsize=(8, 6))
+    plt.scatter(test_df['Age'], test_df['Predicted_Age'], color=colors, label='Predicted vs Actual')
+    # plt.plot(test_df['Age'], test_df['Age'], color='red', linestyle='--', label='Perfect Prediction')  # Optional: Line of perfect prediction
+    # Add legend for colors based on ImageID groups
+    handles = [plt.Line2D([0], [0], marker='o', color=cmap(i), linestyle='', markersize=10) 
+            for i, group in enumerate(unique_groups)]
+    plt.legend(handles, unique_groups, title="ImageID Groups")
+    plt.xlabel('Age')
+    plt.ylabel('Predicted Age')
+    plt.title('Age vs Predicted Age')
+    plt.grid(True)
+    plt.savefig(f"model_dumps/mix/plots/vit_cnn_{sys.argv[1]}_epoch{epoch}.png")
diff --git a/vit_train_feature_cnn_main_mix.py b/vit_train_feature_cnn_main_mix.py
new file mode 100644
index 0000000..a4841a0
--- /dev/null
+++ b/vit_train_feature_cnn_main_mix.py
@@ -0,0 +1,1186 @@
+import os
+import pandas as pd
+import nibabel as nib
+import numpy as np
+import torch
+import torch.nn as nn
+import torch.nn.init as init
+import torch.optim as optim
+from tqdm import tqdm
+from nibabel.orientations import io_orientation, axcodes2ornt, ornt_transform, apply_orientation
+from torchvision import transforms
+from transformers import ViTFeatureExtractor, ViTModel
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler
+import pickle
+import matplotlib.pyplot as plt
+import sys
+import SimpleITK as sitk
+from scipy.ndimage import zoom
+from dataset_cls import ADNIDataset, ADNIDatasetViT
+from torch.utils.data import DataLoader, Dataset
+
+def set_random_seed(seed=69420):
+    np.random.seed(seed)
+    torch.manual_seed(seed)
+    torch.cuda.manual_seed(seed)
+    torch.cuda.manual_seed_all(seed)  # For multi-GPU setups
+    torch.backends.cudnn.deterministic = True
+    torch.backends.cudnn.benchmark = False
+
+universal_seed = 69420
+
+set_random_seed(universal_seed)
+
+# Function to resample the volume to 160 slices
+def resample_volume_to_fixed_slices(data, affine, target_slices=160):
+    # Convert Numpy array and affine to SimpleITK image
+    sitk_img = sitk.GetImageFromArray(data)
+    sitk_img.SetSpacing([affine[0, 0], affine[1, 1], affine[2, 2]])
+
+    original_size = sitk_img.GetSize()  # (width, height, depth)
+    original_spacing = sitk_img.GetSpacing()  # (spacing_x, spacing_y, spacing_z)
+
+    # Calculate new spacing to achieve the target number of slices
+    new_spacing = list(original_spacing)
+    new_spacing[2] = (original_spacing[2] * original_size[2]) / target_slices
+
+    # Define new size
+    new_size = [original_size[0], original_size[1], target_slices]
+
+    # Resample the image
+    resampler = sitk.ResampleImageFilter()
+    resampler.SetOutputSpacing(new_spacing)
+    resampler.SetSize(new_size)
+    resampler.SetInterpolator(sitk.sitkLinear)
+    resampled_img = resampler.Execute(sitk_img)
+
+    return sitk.GetArrayFromImage(resampled_img)  # Return the resampled image as a numpy array
+
+
+def resample_nifti(img_data, target_slices = 160):
+    # Determine the current number of slices along the z-axis (3rd dimension)
+    current_slices = img_data.shape[0]
+    # Calculate the zoom factor for resampling (only along the z-axis)
+    zoom_factor = target_slices / current_slices
+    # Resample the image data along the z-axis
+    resampled_data = zoom(img_data, (zoom_factor, 1, 1), order=3)  # order=3 for cubic interpolation
+    # Ensure that the resampled data has the target number of slices
+    # print (resampled_data.shape)
+    # resampled_data = resampled_data[:target_slices,:,:]
+    # print (resampled_data.shape)
+    return resampled_data
+
+
+# Check if GPU is available
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+# Load the CSV file into a pandas DataFrame
+csv_path = "adni_storage/adni_brainrotnet_metadata.csv"
+df_adni = pd.read_csv(csv_path)
+# df = df.sample(n=1000, random_state=69420)
+# Add a new column 'filepath' with the constructed file paths
+df_adni['filepath'] = df_adni.apply(
+    lambda row: f"adni_storage/ADNI_nii_gz_bias_corrected/I{row['ImageID'][4:]}_{row['SubjectID']}.stripped.N4.nii.gz",
+    axis=1
+)
+df_adni = df_adni.sort_values(by='Age', ascending=True).reset_index(drop=True).head(500)
+# df_adni=df_adni.sample(n=400)
+
+# Load independent dataset metadata
+metadata_path = "ixi_storage/ixi_brainrotnet_metadata.csv"
+df_ixi = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_ixi['filepath'] = df_ixi.apply(
+    lambda row: f"ixi_storage/IXI_bias_corrected/{row['ImageID']}.stripped.N4.nii.gz",
+    axis=1
+)
+
+metadata_path = "abide_storage/abide_brainrotnet_metadata.csv"
+df_abide = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_abide['filepath'] = df_abide.apply(
+    lambda row: f"abide_storage/ABIDEII_bias_corrected/{row['ImageID'][7:]}.stripped.N4.nii.gz",
+    axis=1
+)
+df_abide = df_abide.sort_values(by='Age', ascending=False).reset_index(drop=True).head(500)
+# df_abide=df_abide.sample(n=400)
+
+metadata_path = "dlbs_storage/dlbs_brainrotnet_metadata.csv"
+df_dlbs = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_dlbs['filepath'] = df_dlbs.apply(
+    lambda row: f"dlbs_storage/DLBS_bias_corrected/{row['ImageID'][4:]}.stripped.N4.nii.gz",
+    axis=1
+)
+
+metadata_path = "cobre_storage/cobre_brainrotnet_metadata.csv"
+df_cobre = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_cobre['filepath'] = df_cobre.apply(
+    lambda row: f"cobre_storage/COBRE_bias_corrected/{row['ImageID'][5:]}.stripped.N4.nii.gz",
+    axis=1
+)
+
+metadata_path = "fcon1000_storage/fcon1000_brainrotnet_metadata.csv"
+df_fcon = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_fcon['filepath'] = df_fcon.apply(
+    lambda row: f"fcon1000_storage/fcon1000_bias_corrected/{row['ImageID'][8:]}.stripped.N4.nii.gz",
+    axis=1
+)
+df_fcon = df_fcon.dropna()
+# df_fcon = df_fcon.sort_values(by='Age', ascending=False).reset_index(drop=True).head(300)
+# df_fcon = df_fcon.sample(n=300)
+
+metadata_path = "sald_storage/sald_brainrotnet_metadata.csv"
+df_sald = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_sald['filepath'] = df_sald.apply(
+    lambda row: f"sald_storage/SALD_bias_corrected/sub-{row['ImageID'][4:]}.stripped.N4.nii.gz",
+    axis=1
+)
+# df_sald = df_sald.sort_values(by='Age', ascending=False).reset_index(drop=True).head(300)
+# # df_sald = df_sald.sample(n=300)
+
+metadata_path = "corr_storage/corr_brainrotnet_metadata.csv"
+df_corr = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_corr['filepath'] = df_corr.apply(
+    lambda row: f"corr_storage/CORR_bias_corrected/{row['ImageID'][5:]}.stripped.N4.nii.gz",
+    axis=1
+)
+df_corr = df_corr.sort_values(by='Age', ascending=True).reset_index(drop=True).head(300)
+df_corr = df_corr.sample(n=200)
+
+
+metadata_path = "oasis1_storage/oasis1_brainrotnet_metadata.csv"
+df_oas1 = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_oas1['filepath'] = df_oas1.apply(
+    lambda row: f"oasis1_storage/oasis_nii_gz_bias_corrected/{row['ImageID']}.stripped.N4.nii.gz",
+    axis=1
+)
+df_oas1 = df_oas1.sort_values(by='Age', ascending=False).reset_index(drop=True).head(300)
+
+
+df = pd.concat ([
+                 df_adni[['ImageID', 'Sex', 'Age', 'filepath']], 
+                 df_ixi[['ImageID', 'Sex', 'Age', 'filepath']], 
+                #  df_abide[['ImageID', 'Sex', 'Age', 'filepath']],
+                 df_dlbs[['ImageID', 'Sex', 'Age', 'filepath']],
+                #  df_cobre[['ImageID', 'Sex', 'Age', 'filepath']],
+                 df_fcon[['ImageID', 'Sex', 'Age', 'filepath']],
+                #  df_sald[['ImageID', 'Sex', 'Age', 'filepath']],
+                #  df_corr[['ImageID', 'Sex', 'Age', 'filepath']], 
+                #  df_oas1[['ImageID', 'Sex', 'Age', 'filepath']],
+                 ], ignore_index=True)
+print (df)
+# Ensure 'Age' is an integer
+df['Age_Group'] = df['Age'].astype(int).apply(lambda x: f"{x:03d}"[:-1] + "0")
+df['Age_Group'] = df['Age_Group'] + df['Sex']
+print (df['Age_Group'].unique())
+# Prepare dataset and dataloaders
+sex_encoded = df['Sex'].apply(lambda x: 0 if x == 'M' else 1).tolist()
+age_list = df['Age'].tolist()
+filepath_list = df['filepath'].tolist()
+label_list = df['Age_Group'].tolist()
+
+# Get unique labels and create a mapping
+unique_labels = sorted(set(label_list))  # Ensure consistent ordering
+label_to_idx = {label: idx for idx, label in enumerate(unique_labels)}
+idx_to_label = {idx: label for label, idx in label_to_idx.items()}  # Reverse mapping for decoding
+
+# Convert labels to integers
+numeric_labels = [label_to_idx[label] for label in label_list]
+label_list = numeric_labels
+
+roi = 160
+
+# Transformation pipeline for ViT
+transform = transforms.Compose([
+    transforms.ToPILImage(),
+    transforms.Resize((224, 224)),
+    transforms.Lambda(lambda img: img.convert("RGB")),  # Convert to RGB
+    transforms.ToTensor(),
+    transforms.Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5]),  # Normalize for ViT
+])
+
+
+# Function to extract 16 evenly spaced slices
+def extract_slices(volume, num_slices=16):
+    total_slices = volume.shape[0]
+    indices = np.linspace(0, total_slices - 1, num_slices, dtype=int)
+    return volume[indices, :, :]  # Select slices
+
+# Create ADNIDataset
+vit_dataset = ADNIDatasetViT(filepath_list, label_list)
+
+train_size = int(0.8 * len(vit_dataset))
+val_size = len(vit_dataset) - train_size
+generator = torch.Generator().manual_seed(universal_seed)
+vit_train_dataset, vit_val_dataset = torch.utils.data.random_split(vit_dataset, [train_size, val_size], generator=generator)
+
+print(len(vit_train_dataset))
+
+# DataLoader for training
+vit_train_loader = DataLoader(vit_train_dataset, batch_size=8, shuffle=False)
+
+# Function to preprocess data and dynamically expand slices while saving to disk
+def preprocess_and_expand(dataset, transform, output_dir, num_slices=16):
+    os.makedirs(output_dir, exist_ok=True)  # Ensure output directory exists
+    expanded_images, expanded_labels = [], []
+
+    for filepath, label in tqdm(dataset, desc="Processing Slices"):
+        # Check if all slice files already exist
+        all_slices_exist = True
+        slice_filenames = [
+            os.path.join(output_dir, f"{os.path.basename(filepath)}_slice_{i}.pt")
+            for i in range(num_slices)
+        ]
+        if not all(os.path.exists(slice_file) for slice_file in slice_filenames):
+            all_slices_exist = False
+
+        # Skip processing if all slices exist
+        if all_slices_exist:
+            expanded_images.extend(slice_filenames)  # Add existing file paths
+            expanded_labels.extend([label] * num_slices)
+            continue
+
+        # Load NIfTI image only if slices are missing
+        nii_img = nib.load(filepath)
+        data = nii_img.get_fdata()
+
+        # Normalize and extract slices
+        data = (data - data.min()) / (data.max() - data.min())
+        slices = extract_slices(data, num_slices)
+
+        # Transform each slice, save to file, and add to dataset
+        for i, slice_data in enumerate(slices):
+            slice_filename = slice_filenames[i]
+            if not os.path.exists(slice_filename):
+                transformed_slice = transform(slice_data)  # Transform slice
+                torch.save(transformed_slice, slice_filename)  # Save to file
+            expanded_images.append(slice_filename)  # Store file path
+            expanded_labels.append(label)
+
+    return expanded_images, expanded_labels
+
+# Instantiate Dataset
+vit_dataset = ADNIDatasetViT(filepath_list, label_list)
+
+# Split Dataset
+train_size = int(0.8 * len(vit_dataset))
+val_size = len(vit_dataset) - train_size
+generator = torch.Generator().manual_seed(universal_seed)
+vit_train_dataset, vit_val_dataset = torch.utils.data.random_split(vit_dataset, [train_size, val_size], generator=generator)
+
+# Create New Dataset with Filepaths
+class ExpandedDataset(Dataset):
+    def __init__(self, image_paths, labels):
+        self.image_paths = image_paths
+        self.labels = labels
+
+    def __len__(self):
+        return len(self.labels)
+
+    def __getitem__(self, idx):
+        # Load the image from file
+        image = torch.load(self.image_paths[idx])
+        label = torch.tensor(self.labels[idx], dtype=torch.long)
+        return image, label
+
+# Define output directory for slices
+output_dir = "processed_slices"
+
+# Preprocess and expand the training data
+expanded_image_paths, expanded_labels = preprocess_and_expand(vit_train_dataset, transform, output_dir)
+
+# Create Expanded Dataset and DataLoader
+expanded_train_dataset = ExpandedDataset(expanded_image_paths, expanded_labels)
+expanded_train_loader = DataLoader(expanded_train_dataset, batch_size=8, shuffle=True)
+
+# Print Sizes
+print(f"Original Training Dataset Size: {len(vit_train_dataset)}")
+print(f"Expanded Training Dataset Size: {len(expanded_train_dataset)}")
+
+from transformers import ViTForImageClassification
+# Load ViT model
+num_classes = df['Age_Group'].nunique()  # Number of unique Age_Groups
+model = ViTForImageClassification.from_pretrained(
+    "google/vit-base-patch16-224",
+    num_labels=num_classes,
+    ignore_mismatched_sizes=True, 
+)
+
+model.to(device)
+
+# Loss function and optimizer
+criterion = nn.CrossEntropyLoss()
+optimizer = optim.AdamW(model.parameters(), lr=1e-4)
+
+# Function to save checkpoint
+def save_checkpoint(epoch, model, optimizer, path="model_dumps/vit_train_checkpoint.pth"):
+    torch.save({
+        'epoch': epoch,
+        'model_state_dict': model.state_dict(),
+        'optimizer_state_dict': optimizer.state_dict(),
+    }, path)
+    print(f"Checkpoint saved at epoch {epoch+1}")
+
+# Function to load checkpoint
+def load_checkpoint(path="model_dumps/vit_train_checkpoint.pth"):
+    checkpoint = torch.load(path)
+    model.load_state_dict(checkpoint['model_state_dict'])
+    optimizer.load_state_dict(checkpoint['optimizer_state_dict'])
+    start_epoch = checkpoint['epoch'] + 1
+    print(f"Checkpoint loaded. Resuming from epoch {start_epoch}")
+    return start_epoch
+
+# Check if recovery mode is enabled
+checkpoint_path = "model_dumps/vit_train_checkpoint.pth"
+start_epoch = 0
+
+if len(sys.argv) > 4 and sys.argv[4] == "recover":
+    start_epoch = load_checkpoint(path=checkpoint_path)
+
+# Training loop
+vit_train_epochs = 15
+model.train()
+
+for epoch in range(start_epoch, vit_train_epochs):
+    running_loss = 0.0
+    correct = 0
+    total = 0
+
+    for inputs, labels in tqdm(expanded_train_loader, desc=f"Epoch {epoch+1}/{vit_train_epochs}"):
+        # Move data to device
+        inputs, labels = inputs.to(device), labels.to(device)
+
+        # Forward pass
+        outputs = model(pixel_values=inputs)  # ViT expects `pixel_values`
+        loss = criterion(outputs.logits, labels)
+
+        # Backward pass and optimization
+        optimizer.zero_grad()
+        loss.backward()
+        optimizer.step()
+
+        running_loss += loss.item()
+
+        # Calculate accuracy
+        _, predicted = torch.max(outputs.logits, dim=1)
+        total += labels.size(0)
+        correct += (predicted == labels).sum().item()
+
+        # batch_accuracy = 100 * (predicted == labels).sum().item() / labels.size(0)
+        # print(f"Batch Accuracy: {batch_accuracy:.2f}%")
+
+    epoch_loss = running_loss / len(expanded_train_loader)
+    epoch_accuracy = 100 * correct / total
+    print(f"Epoch {epoch+1}/{vit_train_epochs}, Loss: {epoch_loss:.4f}, Accuracy: {epoch_accuracy:.2f}%")
+
+    # Save checkpoint at the end of each epoch
+    save_checkpoint(epoch, model, optimizer, path=f"model_dumps/vit_train_checkpoint.pth")
+# Load pre-trained ViT model
+feature_extractor = ViTFeatureExtractor.from_pretrained("google/vit-base-patch16-224")
+# model = ViTModel.from_pretrained("google/vit-base-patch16-224")
+# model.to(device)  # Move the model to the GPU (if available)
+model.eval()
+
+# Update image transform for grayscale images to match ViT input requirements
+transform = transforms.Compose([
+    transforms.ToPILImage(),
+    transforms.Resize((224, 224)),
+    transforms.Lambda(lambda img: img.convert("RGB")),  # Convert to RGB directly
+    transforms.ToTensor(),
+    transforms.Normalize(mean=feature_extractor.image_mean, std=feature_extractor.image_std),
+])
+
+torch.cuda.empty_cache()  # Free GPU memory
+
+
+
+# To store features and labels
+features_list = []
+labels_list = []
+
+
+# Directory to save processed images and features
+os.makedirs(f"adni_storage/ADNI_features/train_e{vit_train_epochs}/", exist_ok=True)
+# Process each row in the DataFrame
+for _, row in tqdm(df_adni.iterrows(), total=len(df_adni), desc="Processing images"):
+    filepath = row['filepath']
+    image_title = f"{row['ImageID'][4:]}_{row['SubjectID']}"
+
+    # Check if the feature file already exists
+    feature_file_path = f"adni_storage/ADNI_features/train_e{vit_train_epochs}/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        
+        from PIL import Image
+        # Normalize the array to 0-255 for grayscale image
+        data_normalized = ((features - np.min(features)) / (np.max(features) - np.min(features)) * 255).astype(np.uint8)
+        data_normalized = np.repeat(data_normalized, 4, axis=0)
+        # Create an image from the array
+        img = Image.fromarray(np.transpose(data_normalized), mode='L')  # 'L' mode for grayscale
+        # Save the image
+        # img.save(f"adni_storage/ADNI_features/train_e{vit_train_epochs}/featuremaps/{image_title}_fm.png")
+
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        # print ("hiii")
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_volume_to_fixed_slices(data, affine, target_slices=160)
+
+                # Extract features for all sagittal slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+
+                    # Transform slice for ViT input
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)  # Add batch dimension and move to GPU
+
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        # #outputs = model(slice_tensor)
+                        # slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()  # Move output back to CPU
+                        slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Target is 'Age'
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+
+# Directory to save processed images and features
+os.makedirs(f"ixi_storage/IXI_features/train_e{vit_train_epochs}/", exist_ok=True)
+for _, row in tqdm(df_ixi.iterrows(), total=len(df_ixi), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID']}"
+        # Check if the feature file already exists
+    feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        #outputs = model(slice_tensor)
+                        slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+
+# # Directory to save processed images and features
+# os.makedirs(f"abide_storage/ABIDEII_features/train_e{vit_train_epochs}/", exist_ok=True)
+# for _, row in tqdm(df_abide.iterrows(), total=len(df_abide), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][7:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"abide_storage/ABIDEII_features/train_e{vit_train_epochs}/{image_title}_features.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         #outputs = model(slice_tensor)
+#                         slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+os.makedirs(f"dlbs_storage/DLBS_features/train_e{vit_train_epochs}/", exist_ok=True)
+for _, row in tqdm(df_dlbs.iterrows(), total=len(df_dlbs), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID'][4:]}"
+        # Check if the feature file already exists
+    feature_file_path = f"dlbs_storage/DLBS_features/train_e{vit_train_epochs}/{image_title}_features.npy"
+    # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        #outputs = model(slice_tensor)
+                        slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+
+# os.makedirs(f"cobre_storage/COBRE_features/train_e{vit_train_epochs}/", exist_ok=True)
+# for _, row in tqdm(df_cobre.iterrows(), total=len(df_cobre), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][5:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"cobre_storage/COBRE_features/train_e{vit_train_epochs}/{image_title}_features.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         #outputs = model(slice_tensor)
+#                         slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+os.makedirs(f"fcon1000_storage/fcon1000_features/train_e{vit_train_epochs}/", exist_ok=True)
+for _, row in tqdm(df_fcon.iterrows(), total=len(df_fcon), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID'][5:]}"
+        # Check if the feature file already exists
+    feature_file_path = f"fcon1000_storage/fcon1000_features/train_e{vit_train_epochs}/{image_title}_features.npy"
+    # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        #outputs = model(slice_tensor)
+                        slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+# os.makedirs(f"sald_storage/SALD_features/train_e{vit_train_epochs}/", exist_ok=True)
+# for _, row in tqdm(df_sald.iterrows(), total=len(df_sald), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][4:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"sald_storage/SALD_features/train_e{vit_train_epochs}/{image_title}_features.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         #outputs = model(slice_tensor)
+#                         slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+# os.makedirs(f"corr_storage/CORR_features/train_e{vit_train_epochs}/", exist_ok=True)
+# for _, row in tqdm(df_corr.iterrows(), total=len(df_corr), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][5:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"corr_storage/CORR_features/train_e{vit_train_epochs}/{image_title}_features.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         #outputs = model(slice_tensor)
+#                         slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+
+# os.makedirs(f"oasis1_storage/oasis1_features/train_e{vit_train_epochs}/", exist_ok=True)
+# for _, row in tqdm(df_oas1.iterrows(), total=len(df_oas1), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID']}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"oasis1_storage/oasis1_features/train_e{vit_train_epochs}/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         #outputs = model(slice_tensor)
+#                         slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+
+
+batch_size = 1
+
+# print (features_list)
+print (features_list[0].shape)
+
+# Create Dataset and DataLoader
+dataset = ADNIDataset(features_list, sex_encoded, age_list)
+train_size = int(0.8 * len(dataset))
+val_size = len(dataset) - train_size
+generator.manual_seed(universal_seed)
+train_dataset, val_dataset = torch.utils.data.random_split(dataset, [train_size, val_size])
+# Store the indices of the validation dataset
+val_indices = val_dataset.indices
+train_indices = train_dataset.indices
+
+train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=False)
+val_loader = DataLoader(val_dataset, batch_size=batch_size, shuffle=False)
+
+# Tracking outputs for validation samples
+val_outputs = {}
+train_outputs = {}
+
+# Initialize model, loss, and optimizer
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+
+###########################################
+# THIS IS WHERE YOU CHOOSE A MODEL TO TEST 
+###########################################
+
+import importlib
+
+# Assuming sys.argv[1] is the module name
+module_name = sys.argv[1]  # Example: "my_model"
+class_name = "AgePredictionCNN"  # The class you want to import
+
+try:
+    # Dynamically import the module
+    module = importlib.import_module(module_name)
+    
+    # Dynamically get the class
+    AgePredictionCNN = getattr(module, class_name)
+    
+    print(f"Successfully imported {class_name} from {module_name}.")
+
+except ImportError:
+    print(f"Module {module_name} could not be imported.")
+except AttributeError:
+    print(f"{class_name} does not exist in {module_name}.")
+
+##############################
+# MODEL IMPORTED DYNAMICALLY
+##############################
+
+print (features_list[0].shape)
+model = AgePredictionCNN((1, features_list[0].shape[0], features_list[0].shape[1])).to(device)
+criterion = nn.MSELoss()  # MAE Loss
+eval_crit = nn.L1Loss()
+optimizer = optim.Adam(model.parameters(), lr=0.001)
+best_loss = np.inf  # Initialize the best loss to infinity
+start_epoch = 0
+
+
+load_saved = sys.argv[2] # "last, "best"
+if load_saved != "none":
+    with open(f"model_dumps/mix/{sys.argv[1]}_best_model_with_metadata.pkl", "rb") as f:
+        checkpoint = pickle.load(f)
+    best_loss = checkpoint["loss"]
+
+    # Load the checkpoint
+    with open(f"model_dumps/mix/{sys.argv[1]}_{load_saved}_model_with_metadata.pkl", "rb") as f:
+        checkpoint = pickle.load(f)
+
+    # Restore model and optimizer state
+    model.load_state_dict(checkpoint["model_state"])
+    optimizer.load_state_dict(checkpoint["optimizer_state"])
+
+    # Restore RNG states
+    torch.set_rng_state(checkpoint["t_rng_st"])
+    np.random.set_state(checkpoint["n_rng_st"])
+    if torch.cuda.is_available() and checkpoint["cuda_rng_st"] is not None:
+        torch.cuda.set_rng_state_all(checkpoint["cuda_rng_st"])
+
+    # Retrieve metadata
+    start_epoch = checkpoint["epoch"] + 1
+    loaded_loss = checkpoint["loss"]
+
+    print(f"Loaded model from epoch {start_epoch} with validation loss {loaded_loss:.4f}, best loss {best_loss:.4f}")
+
+predicted_ages = None
+# Training loop
+epochs = int(sys.argv[3])
+
+# Initialize lists to track loss
+filename = sys.argv[1] 
+csv_file = f"model_dumps/mix/{filename}.csv"
+
+# Load existing epoch data if the file exists
+if os.path.exists(csv_file):
+    epoch_data = pd.read_csv(csv_file).to_dict(orient="records")
+    print(f"Loaded existing epoch data from {csv_file}.")
+else:
+    epoch_data = []
+    print("No existing epoch data found. Starting fresh.")
+
+
+# Plot loss vs. epoch and save the figure
+def update_loss_plot(epoch_data, filename):
+    df = pd.DataFrame(epoch_data)
+    df.to_csv(f"model_dumps/mix/{filename}.csv", index=False)  # Save the data to CSV
+    
+    plt.figure(figsize=(8, 6))
+    plt.plot(df['epoch'], df['train_loss'], label="Train Loss", marker="o")
+    plt.plot(df['epoch'], df['val_loss'], label="Validation Loss", marker="o")
+    plt.xlabel("Epoch")
+    plt.ylabel("Loss")
+    plt.title("Loss vs. Epoch")
+    plt.legend()
+    plt.grid(True)
+    plt.savefig(f"model_dumps/mix/{filename}.png")
+    plt.close()
+
+# Training loop
+for epoch in range(start_epoch, epochs):
+    model.train()
+    train_loss = 0.0
+    predicted_ages = []
+
+    for idx, (features, sex, age) in enumerate(train_loader):
+        features = features.unsqueeze(1).to(device)  # Add channel dimension
+        sex = sex.to(device)
+        age = age.to(device)
+        optimizer.zero_grad()
+        outputs = model(features, sex)
+        # Store the output for each sample in the batch
+        for i in range(outputs.size(0)):
+            train_outputs[train_indices[idx * batch_size + i]] = outputs[i].item()
+
+        loss = criterion(outputs.squeeze(), age)
+        loss.backward()
+        optimizer.step()
+
+        train_loss += loss.item()
+
+    train_loss /= len(train_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Train Loss: {train_loss:.4f}")
+
+    # Validation loop
+    model.eval()
+    val_loss = 0.0
+    with torch.no_grad():
+        for idx, (features, sex, age) in enumerate(val_loader):
+            features = features.unsqueeze(1).to(device)
+            sex = sex.to(device)
+            age = age.to(device)
+
+            outputs = model(features, sex)
+            loss = eval_crit(outputs.squeeze(), age)
+            val_loss += loss.item()
+
+            # Save the predicted age for the current validation sample
+            for i in range(outputs.size(0)):
+                val_outputs[val_indices[idx * batch_size + i]] = outputs[i].item()
+            # val_outputs[val_indices[idx]] = outputs.item()
+
+    val_loss /= len(val_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Validation Loss: {val_loss:.4f}")
+
+    # Save the last model with metadata
+    print(f"Saving last model...")
+    checkpoint = {
+        "model_state": model.state_dict(),
+        "optimizer_state": optimizer.state_dict(),
+        "epoch": epoch,
+        "loss": val_loss,
+        "t_rng_st": torch.get_rng_state(),
+        "n_rng_st": np.random.get_state(),
+        "cuda_rng_st": torch.cuda.get_rng_state_all() if torch.cuda.is_available() else None
+    }
+    with open(f"model_dumps/mix/{sys.argv[1]}_last_model_with_metadata.pkl", "wb") as f:
+        pickle.dump(checkpoint, f)
+    print(f"Last model saved...")
+
+    # Check if validation loss improved
+    if val_loss < best_loss:
+        best_loss = val_loss
+        print(f"Validation loss improved to {best_loss:.4f}. Saving model...")
+        with open(f"model_dumps/mix/{sys.argv[1]}_best_model_with_metadata.pkl", "wb") as f:
+            pickle.dump(checkpoint, f)
+
+    # Save predictions and create DataFrames (same as before)
+    # ...
+
+    # Update epoch data and save the loss plot
+    epoch_data.append({
+        "epoch": epoch + 1,
+        "train_loss": train_loss,
+        "val_loss": val_loss
+    })
+    update_loss_plot(epoch_data, sys.argv[1])
+
+    max_index = max(train_outputs.keys())
+    # Create a DataFrame with NaN for all indices initially
+    df_trn = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
+    # Assign the values to their respective indices
+    for index, value in train_outputs.items():
+        df_trn.loc[index, "Predicted_Age"] = value
+    # print (df_trn)
+
+    df2 = df.copy()
+    df2['Predicted_Age'] = df_trn['Predicted_Age']
+    train_df = df2.loc[train_outputs.keys()]
+    # print (train_df)
+    train_df.to_csv(f"model_dumps/mix/{sys.argv[1]}_predicted_ages_train.csv")
+
+    max_index = max(val_outputs.keys())
+    # Create a DataFrame with NaN for all indices initially
+    df_pred = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
+    # Assign the values to their respective indices
+    for index, value in val_outputs.items():
+        df_pred.loc[index, "Predicted_Age"] = value
+    # print (df_pred)
+
+    df1 = df.copy()
+    df1['Predicted_Age'] = df_pred['Predicted_Age']
+    test_df = df1.loc[val_outputs.keys()]
+    print (test_df)
+    test_df.to_csv(f"model_dumps/mix/{sys.argv[1]}_predicted_ages_val.csv")
+
+
+    # Map unique first 4 characters of ImageID to color codes
+    unique_groups = test_df['ImageID'].str[:3].unique()
+    group_to_color = {group: i for i, group in enumerate(unique_groups)}
+
+    # Assign colors based on the mapping
+    cmap = plt.get_cmap('tab10')  # Change colormap as desired
+    colors = [cmap(group_to_color[group]) for group in test_df['ImageID'].str[:3]]
+
+    # Check that the predictions have been added to the DataFrame
+    # Plot Age vs. Predicted Age
+    plt.figure(figsize=(8, 6))
+    plt.scatter(test_df['Age'], test_df['Predicted_Age'], color=colors, label='Predicted vs Actual')
+    # plt.plot(test_df['Age'], test_df['Age'], color='red', linestyle='--', label='Perfect Prediction')  # Optional: Line of perfect prediction
+    # Add legend for colors based on ImageID groups
+    handles = [plt.Line2D([0], [0], marker='o', color=cmap(i), linestyle='', markersize=10) 
+            for i, group in enumerate(unique_groups)]
+    plt.legend(handles, unique_groups, title="ImageID Groups")
+    plt.xlabel('Age')
+    plt.ylabel('Predicted Age')
+    plt.title('Age vs Predicted Age')
+    plt.grid(True)
+    plt.savefig(f"model_dumps/mix/plots/vit_cnn_{sys.argv[1]}_epoch{epoch}.png")
diff --git a/vit_train_feature_cnn_main_mix_roi.py b/vit_train_feature_cnn_main_mix_roi.py
new file mode 100644
index 0000000..c54e4ad
--- /dev/null
+++ b/vit_train_feature_cnn_main_mix_roi.py
@@ -0,0 +1,1304 @@
+import os
+import pandas as pd
+import nibabel as nib
+import numpy as np
+import torch
+import torch.nn as nn
+import torch.nn.init as init
+import torch.optim as optim
+from tqdm import tqdm
+from nibabel.orientations import io_orientation, axcodes2ornt, ornt_transform, apply_orientation
+from torchvision import transforms
+from transformers import ViTFeatureExtractor, ViTModel
+from transformers import ViTForImageClassification
+from transformers import DeiTForImageClassification
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler
+import pickle
+import matplotlib.pyplot as plt
+import sys
+import SimpleITK as sitk
+from scipy.ndimage import zoom
+from dataset_cls import ADNIDataset, ADNIDatasetViT
+from torch.utils.data import DataLoader, Dataset
+import gc
+
+model_name = "vit"
+
+def set_random_seed(seed=69420):
+    np.random.seed(seed)
+    torch.manual_seed(seed)
+    torch.cuda.manual_seed(seed)
+    torch.cuda.manual_seed_all(seed)  # For multi-GPU setups
+    torch.backends.cudnn.deterministic = True
+    torch.backends.cudnn.benchmark = False
+
+universal_seed = 69420
+
+set_random_seed(universal_seed)
+
+# Function to resample the volume to 160 slices
+def resample_volume_to_fixed_slices(data, affine, target_slices=160):
+    # Convert Numpy array and affine to SimpleITK image
+    sitk_img = sitk.GetImageFromArray(data)
+    sitk_img.SetSpacing([affine[0, 0], affine[1, 1], affine[2, 2]])
+
+    original_size = sitk_img.GetSize()  # (width, height, depth)
+    original_spacing = sitk_img.GetSpacing()  # (spacing_x, spacing_y, spacing_z)
+
+    # Calculate new spacing to achieve the target number of slices
+    new_spacing = list(original_spacing)
+    new_spacing[2] = (original_spacing[2] * original_size[2]) / target_slices
+
+    # Define new size
+    new_size = [original_size[0], original_size[1], target_slices]
+
+    # Resample the image
+    resampler = sitk.ResampleImageFilter()
+    resampler.SetOutputSpacing(new_spacing)
+    resampler.SetSize(new_size)
+    resampler.SetInterpolator(sitk.sitkLinear)
+    resampled_img = resampler.Execute(sitk_img)
+
+    return sitk.GetArrayFromImage(resampled_img)  # Return the resampled image as a numpy array
+
+
+def resample_nifti(img_data, target_slices = 160):
+    # Determine the current number of slices along the z-axis (3rd dimension)
+    current_slices = img_data.shape[0]
+    # Calculate the zoom factor for resampling (only along the z-axis)
+    zoom_factor = target_slices / current_slices
+    # Resample the image data along the z-axis
+    resampled_data = zoom(img_data, (zoom_factor, 1, 1), order=3)  # order=3 for cubic interpolation
+    # Ensure that the resampled data has the target number of slices
+    # print (resampled_data.shape)
+    # resampled_data = resampled_data[:target_slices,:,:]
+    # print (resampled_data.shape)
+    return resampled_data
+
+
+# Check if GPU is available
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+# Load the CSV file into a pandas DataFrame
+csv_path = "adni_storage/adni_brainrotnet_metadata.csv"
+df_adni = pd.read_csv(csv_path)
+# df = df.sample(n=1000, random_state=69420)
+# Add a new column 'filepath' with the constructed file paths
+df_adni['filepath'] = df_adni.apply(
+    lambda row: f"adni_storage/ADNI_nii_gz_bias_corrected/I{row['ImageID'][4:]}_{row['SubjectID']}.stripped.N4.nii.gz",
+    axis=1
+)
+df_adni = df_adni.sort_values(by='Age', ascending=True).reset_index(drop=True).head(500)
+# df_adni=df_adni.sample(n=400)
+
+# Load independent dataset metadata
+# metadata_path = "ixi_storage/ixi_brainrotnet_metadata.csv"
+# df_ixi = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_ixi['filepath'] = df_ixi.apply(
+#     lambda row: f"ixi_storage/IXI_bias_corrected/{row['ImageID']}.stripped.N4.nii.gz",
+#     axis=1
+# )
+
+# metadata_path = "abide_storage/abide_brainrotnet_metadata.csv"
+# df_abide = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_abide['filepath'] = df_abide.apply(
+#     lambda row: f"abide_storage/ABIDEII_bias_corrected/{row['ImageID'][7:]}.stripped.N4.nii.gz",
+#     axis=1
+# )
+# df_abide = df_abide.sort_values(by='Age', ascending=False).reset_index(drop=True).head(500)
+# df_abide=df_abide.sample(n=400)
+
+metadata_path = "dlbs_storage/dlbs_brainrotnet_metadata.csv"
+df_dlbs = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_dlbs['filepath'] = df_dlbs.apply(
+    lambda row: f"dlbs_storage/DLBS_bias_corrected/{row['ImageID'][4:]}.stripped.N4.nii.gz",
+    axis=1
+)
+
+# metadata_path = "cobre_storage/cobre_brainrotnet_metadata.csv"
+# df_cobre = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_cobre['filepath'] = df_cobre.apply(
+#     lambda row: f"cobre_storage/COBRE_bias_corrected/{row['ImageID'][5:]}.stripped.N4.nii.gz",
+#     axis=1
+# )
+
+metadata_path = "fcon1000_storage/fcon1000_brainrotnet_metadata.csv"
+df_fcon = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_fcon['filepath'] = df_fcon.apply(
+    lambda row: f"fcon1000_storage/fcon1000_bias_corrected/{row['ImageID'][8:]}.stripped.N4.nii.gz",
+    axis=1
+)
+df_fcon = df_fcon.dropna()
+# df_fcon = df_fcon.sort_values(by='Age', ascending=False).reset_index(drop=True).head(300)
+# df_fcon = df_fcon.sample(n=300)
+
+metadata_path = "sald_storage/sald_brainrotnet_metadata.csv"
+df_sald = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_sald['filepath'] = df_sald.apply(
+    lambda row: f"sald_storage/SALD_bias_corrected/sub-{row['ImageID'][4:]}.stripped.N4.nii.gz",
+    axis=1
+)
+df_sald = df_sald.sort_values(by='Age', ascending=False).reset_index(drop=True).head(300)
+# df_sald = df_sald.sample(n=300)
+
+# metadata_path = "corr_storage/corr_brainrotnet_metadata.csv"
+# df_corr = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_corr['filepath'] = df_corr.apply(
+#     lambda row: f"corr_storage/CORR_bias_corrected/{row['ImageID'][5:]}.stripped.N4.nii.gz",
+#     axis=1
+# )
+# df_corr = df_corr.sort_values(by='Age', ascending=True).reset_index(drop=True).head(300)
+# df_corr = df_corr.sample(n=200)
+
+
+metadata_path = "oasis1_storage/oasis1_brainrotnet_metadata.csv"
+df_oas1 = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_oas1['filepath'] = df_oas1.apply(
+    lambda row: f"oasis1_storage/oasis_nii_gz_bias_corrected/{row['ImageID']}.stripped.N4.nii.gz",
+    axis=1
+)
+df_oas1 = df_oas1.sort_values(by='Age', ascending=False).reset_index(drop=True).head(300)
+metadata_path = "camcan_storage/camcan_brainrotnet_metadata.csv"
+df_camcan = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_camcan['filepath'] = df_camcan.apply(
+    lambda row: f"camcan_storage/CamCAN_nii_gz_bias_corrected/{row['ImageID']}.stripped.N4.nii.gz",
+    axis=1
+)
+
+
+df = pd.concat ([
+                 df_adni[['ImageID', 'Sex', 'Age', 'filepath']], 
+                #  df_ixi[['ImageID', 'Sex', 'Age', 'filepath']], 
+                #  df_abide[['ImageID', 'Sex', 'Age', 'filepath']],
+                 df_dlbs[['ImageID', 'Sex', 'Age', 'filepath']],
+                #  df_cobre[['ImageID', 'Sex', 'Age', 'filepath']],
+                 df_fcon[['ImageID', 'Sex', 'Age', 'filepath']],
+                 df_sald[['ImageID', 'Sex', 'Age', 'filepath']],
+                #  df_corr[['ImageID', 'Sex', 'Age', 'filepath']], 
+                 df_oas1[['ImageID', 'Sex', 'Age', 'filepath']],
+                df_camcan[['ImageID', 'Sex', 'Age', 'filepath']]
+                 ], ignore_index=True)
+print (df)
+# Ensure 'Age' is an integer
+df['Age_Group'] = df['Age'].astype(int).apply(lambda x: f"{x:03d}"[:-1] + "0")
+df['Age_Group'] = df['Age_Group'] + df['Sex']
+print (df['Age_Group'].unique())
+# Prepare dataset and dataloaders
+sex_encoded = df['Sex'].apply(lambda x: 0 if x == 'M' else 1).tolist()
+age_list = df['Age'].tolist()
+filepath_list = df['filepath'].tolist()
+label_list = df['Age_Group'].tolist()
+
+# Get unique labels and create a mapping
+unique_labels = sorted(set(label_list))  # Ensure consistent ordering
+label_to_idx = {label: idx for idx, label in enumerate(unique_labels)}
+idx_to_label = {idx: label for label, idx in label_to_idx.items()}  # Reverse mapping for decoding
+
+# Convert labels to integers
+numeric_labels = [label_to_idx[label] for label in label_list]
+label_list = numeric_labels
+
+roi = 160
+
+# Transformation pipeline for ViT
+transform = transforms.Compose([
+    transforms.ToPILImage(),
+    transforms.Resize((224, 224)),
+    transforms.Lambda(lambda img: img.convert("RGB")),  # Convert to RGB
+    transforms.ToTensor(),
+    transforms.Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5]),  # Normalize for ViT
+])
+
+
+# Function to extract 16 evenly spaced slices
+def extract_slices(volume, num_slices=16):
+    total_slices = volume.shape[0]
+    indices = np.linspace(0, total_slices - 1, num_slices, dtype=int)
+    return volume[indices, :, :]  # Select slices
+
+def calculate_bounding_box_from_volume(volume):
+    # Find indices of non-zero values
+    non_zero_indices = np.argwhere(volume > 0)
+
+    # Calculate min and max indices along each dimension
+    min_indices = np.min(non_zero_indices, axis=0)
+    max_indices = np.max(non_zero_indices, axis=0)
+
+    # Convert indices to integers
+    min_indices = min_indices.astype(int)
+    max_indices = max_indices.astype(int)
+
+    return min_indices, max_indices
+
+
+def crop_brain_volumes(brain_data):
+    
+
+        # Calculate bounding box from the brain volume
+    min_indices, max_indices = calculate_bounding_box_from_volume(brain_data)
+
+        # Crop the volume
+    cropped_brain = brain_data[min_indices[0]:max_indices[0] + 1,
+                                   min_indices[1]:max_indices[1] + 1,
+                                   min_indices[2]:max_indices[2] + 1]
+    return cropped_brain
+
+# Function to preprocess data and dynamically expand slices while saving to disk
+def preprocess_and_expand(dataset, transform, output_dir, num_slices=16):
+    os.makedirs(output_dir, exist_ok=True)  # Ensure output directory exists
+    expanded_images, expanded_labels = [], []
+
+    for filepath, label in tqdm(dataset, desc="Processing Slices"):
+        # Check if all slice files already exist
+        all_slices_exist = True
+        slice_filenames = [
+            os.path.join(output_dir, f"{os.path.basename(filepath)}_slice_{i}.pt")
+            for i in range(num_slices)
+        ]
+        if not all(os.path.exists(slice_file) for slice_file in slice_filenames):
+            all_slices_exist = False
+
+        # Skip processing if all slices exist
+        if all_slices_exist:
+            expanded_images.extend(slice_filenames)  # Add existing file paths
+            expanded_labels.extend([label] * num_slices)
+            continue
+
+        # Load NIfTI image only if slices are missing
+        nii_img = nib.load(filepath)
+        orig_ornt = io_orientation(nii_img.affine)
+        ras_ornt = axcodes2ornt(("R", "A", "S"))
+        ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+        data = nii_img.get_fdata()  # Load image data
+        data = apply_orientation(data, ornt_trans)
+
+        data = crop_brain_volumes(data)
+
+        # Normalize and extract slices
+        data = (data - data.min()) / (data.max() - data.min())
+        slices = extract_slices(data, num_slices)
+
+        # Transform each slice, save to file, and add to dataset
+        for i, slice_data in enumerate(slices):
+            slice_filename = slice_filenames[i]
+            if not os.path.exists(slice_filename):
+                transformed_slice = transform(slice_data)  # Transform slice
+                torch.save(transformed_slice, slice_filename)  # Save to file
+            expanded_images.append(slice_filename)  # Store file path
+            expanded_labels.append(label)
+
+    return expanded_images, expanded_labels
+# Instantiate Dataset
+vit_dataset = ADNIDatasetViT(filepath_list, label_list)
+
+# Split Dataset
+train_size = int(0.8 * len(vit_dataset))
+val_size = len(vit_dataset) - train_size
+generator = torch.Generator().manual_seed(universal_seed)
+vit_train_dataset, vit_val_dataset = torch.utils.data.random_split(vit_dataset, [train_size, val_size], generator=generator)
+
+# Create New Dataset with Filepaths
+class ExpandedDataset(Dataset):
+    def __init__(self, image_paths, labels):
+        self.image_paths = image_paths
+        self.labels = labels
+
+    def __len__(self):
+        return len(self.labels)
+
+    def __getitem__(self, idx):
+        # Load the image from file
+        image = torch.load(self.image_paths[idx])
+        label = torch.tensor(self.labels[idx], dtype=torch.long)
+        return image, label
+
+slice_count = 32
+
+# Define output directory for slices
+output_dir = f"processed_slices/{slice_count}"
+
+# Preprocess and expand the training data
+expanded_image_paths, expanded_labels = preprocess_and_expand(vit_train_dataset, transform, output_dir, num_slices=slice_count)
+
+# Create Expanded Dataset and DataLoader
+expanded_train_dataset = ExpandedDataset(expanded_image_paths, expanded_labels)
+expanded_train_loader = DataLoader(expanded_train_dataset, batch_size=8, shuffle=True)
+
+# Print Sizes
+print(f"Original Training Dataset Size: {len(vit_train_dataset)}")
+print(f"Expanded Training Dataset Size: {len(expanded_train_dataset)}")
+
+
+# Load ViT model
+num_classes = df['Age_Group'].nunique()  # Number of unique Age_Groups
+model = ViTForImageClassification.from_pretrained(
+    "google/vit-base-patch16-224",
+    num_labels=num_classes,
+    ignore_mismatched_sizes=True, 
+)
+
+
+
+model.to(device)
+
+# Loss function and optimizer
+criterion = nn.CrossEntropyLoss()
+optimizer = optim.AdamW(model.parameters(), lr=1e-4)
+
+# Function to save checkpoint
+def save_checkpoint(epoch, model, optimizer, path=f"model_dumps/{model_name}_train_checkpoint_{slice_count}.pth"):
+    torch.save({
+        'epoch': epoch,
+        'model_state_dict': model.state_dict(),
+        'optimizer_state_dict': optimizer.state_dict(),
+    }, path)
+    print(f"Checkpoint saved at epoch {epoch+1}")
+
+# Function to load checkpoint
+def load_checkpoint(path=f"model_dumps/{model_name}_train_checkpoint_{slice_count}.pth"):
+    checkpoint = torch.load(path)
+    model.load_state_dict(checkpoint['model_state_dict'])
+    optimizer.load_state_dict(checkpoint['optimizer_state_dict'])
+    start_epoch = checkpoint['epoch'] + 1
+    print(f"Checkpoint loaded. Resuming from epoch {start_epoch}")
+    return start_epoch
+
+# Check if recovery mode is enabled
+checkpoint_path = f"model_dumps/{model_name}_train_checkpoint_{slice_count}.pth"
+start_epoch = 0
+
+if len(sys.argv) > 4 and sys.argv[4] == "recover":
+    start_epoch = load_checkpoint(path=checkpoint_path)
+
+# Training loop
+vit_train_epochs = 1
+model.train()
+
+for epoch in range(start_epoch, vit_train_epochs):
+    running_loss = 0.0
+    correct = 0
+    total = 0
+
+    for inputs, labels in tqdm(expanded_train_loader, desc=f"Epoch {epoch+1}/{vit_train_epochs}"):
+        # Move data to device
+        inputs, labels = inputs.to(device), labels.to(device)
+
+        # Forward pass
+        outputs = model(pixel_values=inputs)  # ViT expects `pixel_values`
+        loss = criterion(outputs.logits, labels)
+
+        # Backward pass and optimization
+        optimizer.zero_grad()
+        loss.backward()
+        optimizer.step()
+
+        running_loss += loss.item()
+
+        # Calculate accuracy
+        _, predicted = torch.max(outputs.logits, dim=1)
+        total += labels.size(0)
+        correct += (predicted == labels).sum().item()
+
+        # batch_accuracy = 100 * (predicted == labels).sum().item() / labels.size(0)
+        # print(f"Batch Accuracy: {batch_accuracy:.2f}%")
+
+    epoch_loss = running_loss / len(expanded_train_loader)
+    epoch_accuracy = 100 * correct / total
+    print(f"Epoch {epoch+1}/{vit_train_epochs}, Loss: {epoch_loss:.4f}, Accuracy: {epoch_accuracy:.2f}%")
+    # sav the accuracy and loss for each epoch in a csv file
+    with open(f'{model_name}_train_metrics_{slice_count}.csv', 'a') as f:
+        f.write(f"{epoch+1},{epoch_loss},{epoch_accuracy}\n")
+    # Save checkpoint at the end of each epoch
+    # shift the model to cpu before saving
+    # model.to('cpu')
+    save_checkpoint(epoch, model, optimizer, path=f"model_dumps/{model_name}_train_checkpoint_{slice_count}.pth")
+    # model.to(device)  # Move back to GPU
+    gc.collect()
+
+# Load pre-trained ViT model
+feature_extractor = ViTFeatureExtractor.from_pretrained("google/vit-base-patch16-224")
+# model = ViTModel.from_pretrained("google/vit-base-patch16-224")
+# model.to(device)  # Move the model to the GPU (if available)
+model.eval()
+
+# Update image transform for grayscale images to match ViT input requirements
+transform = transforms.Compose([
+    transforms.ToPILImage(),
+    transforms.Resize((224, 224)),
+    transforms.Lambda(lambda img: img.convert("RGB")),  # Convert to RGB directly
+    transforms.ToTensor(),
+    transforms.Normalize(mean=feature_extractor.image_mean, std=feature_extractor.image_std),
+])
+
+torch.cuda.empty_cache()  # Free GPU memory
+
+
+
+# To store features and labels
+features_list = []
+labels_list = []
+
+
+# Directory to save processed images and features
+os.makedirs(f"adni_storage/ADNI_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+# Process each row in the DataFrame,
+for _, row in tqdm(df_adni.iterrows(), total=len(df_adni), desc="Processing images"):
+    filepath = row['filepath']
+    image_title = f"{row['ImageID'][4:]}_{row['SubjectID']}"
+
+    # Check if the feature file already exists
+    feature_file_path = f"adni_storage/ADNI_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        
+        from PIL import Image
+        # Normalize the array to 0-255 for grayscale image
+        data_normalized = ((features - np.min(features)) / (np.max(features) - np.min(features)) * 255).astype(np.uint8)
+        data_normalized = np.repeat(data_normalized, 4, axis=0)
+        # Create an image from the array
+        img = Image.fromarray(np.transpose(data_normalized), mode='L')  # 'L' mode for grayscale
+        # Save the image
+        # img.save(f"adni_storage/ADNI_features/train_e{vit_train_epochs}/{slice_count}/featuremaps/{image_title}_fm.png")
+
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        # print ("hiii")
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                data = crop_brain_volumes(data)
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+
+                # Extract features for all sagittal slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+
+                    # Transform slice for ViT input
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)  # Add batch dimension and move to GPU
+
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        # #outputs = model(slice_tensor)
+                        # slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()  # Move output back to CPU
+                        slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Target is 'Age'
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+
+# Directory to save processed images and features
+# os.makedirs(f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+# for _, row in tqdm(df_ixi.iterrows(), total=len(df_ixi), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID']}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         #outputs = model(slice_tensor)
+#                         slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+
+# # Directory to save processed images and features
+# os.makedirs(f"abide_storage/ABIDEII_features/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+# for _, row in tqdm(df_abide.iterrows(), total=len(df_abide), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][7:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"abide_storage/ABIDEII_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         #outputs = model(slice_tensor)
+#                         slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+os.makedirs(f"dlbs_storage/DLBS_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+for _, row in tqdm(df_dlbs.iterrows(), total=len(df_dlbs), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID'][4:]}"
+        # Check if the feature file already exists
+    feature_file_path = f"dlbs_storage/DLBS_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+    # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                data = crop_brain_volumes(data)
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        #outputs = model(slice_tensor)
+                        slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+
+# os.makedirs(f"cobre_storage/COBRE_features/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+# for _, row in tqdm(df_cobre.iterrows(), total=len(df_cobre), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][5:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"cobre_storage/COBRE_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         #outputs = model(slice_tensor)
+#                         slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+os.makedirs(f"fcon1000_storage/fcon1000_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+for _, row in tqdm(df_fcon.iterrows(), total=len(df_fcon), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID'][5:]}"
+        # Check if the feature file already exists
+    feature_file_path = f"fcon1000_storage/fcon1000_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+    # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                data = crop_brain_volumes(data)
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        #outputs = model(slice_tensor)
+                        slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+os.makedirs(f"sald_storage/SALD_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+for _, row in tqdm(df_sald.iterrows(), total=len(df_sald), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID'][4:]}"
+        # Check if the feature file already exists
+    feature_file_path = f"sald_storage/SALD_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+    # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        #outputs = model(slice_tensor)
+                        slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+# os.makedirs(f"corr_storage/CORR_features/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+# for _, row in tqdm(df_corr.iterrows(), total=len(df_corr), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][5:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"corr_storage/CORR_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         #outputs = model(slice_tensor)
+#                         slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+
+os.makedirs(f"oasis1_storage/oasis1_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+for _, row in tqdm(df_oas1.iterrows(), total=len(df_oas1), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID']}"
+        # Check if the feature file already exists
+    feature_file_path = f"oasis1_storage/oasis1_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        #outputs = model(slice_tensor)
+                        slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+os.makedirs(f"camcan_storage/CamCAN_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/", exist_ok=True)
+for _, row in tqdm(df_camcan.iterrows(), total=len(df_camcan), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID']}"
+        # Check if the feature file already exists
+    feature_file_path = f"camcan_storage/CamCAN_features/{model_name}/train_e{vit_train_epochs}/{slice_count}/{image_title}_features.npy"
+    # feature_file_path = f"ixi_storage/IXI_features/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                data = crop_brain_volumes(data)
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                      
+                        slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+
+
+batch_size = 1
+
+# print (features_list)
+print (features_list[0].shape)
+
+# Create Dataset and DataLoader
+dataset = ADNIDataset(features_list, sex_encoded, age_list)
+train_size = int(0.8 * len(dataset))
+val_size = len(dataset) - train_size
+generator.manual_seed(universal_seed)
+train_dataset, val_dataset = torch.utils.data.random_split(dataset, [train_size, val_size], generator=generator)
+# Store the indices of the validation dataset
+val_indices = val_dataset.indices
+train_indices = train_dataset.indices
+
+train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=False)
+val_loader = DataLoader(val_dataset, batch_size=batch_size, shuffle=False)
+
+# Tracking outputs for validation samples
+val_outputs = {}
+train_outputs = {}
+
+# Initialize model, loss, and optimizer
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+
+###########################################
+# THIS IS WHERE YOU CHOOSE A MODEL TO TEST 
+###########################################
+
+import importlib
+
+# Assuming sys.argv[1] is the module name
+module_name = sys.argv[1]  # Example: "my_model"
+class_name = "AgePredictionCNN"  # The class you want to import
+
+try:
+    # Dynamically import the module
+    module = importlib.import_module(module_name)
+    
+    # Dynamically get the class
+    AgePredictionCNN = getattr(module, class_name)
+    
+    print(f"Successfully imported {class_name} from {module_name}.")
+
+except ImportError:
+    print(f"Module {module_name} could not be imported.")
+except AttributeError:
+    print(f"{class_name} does not exist in {module_name}.")
+
+##############################
+# MODEL IMPORTED DYNAMICALLY
+##############################
+
+print (features_list[0].shape)
+model = AgePredictionCNN((1, features_list[0].shape[0], features_list[0].shape[1])).to(device)
+criterion = nn.L1Loss()
+
+eval_crit = nn.L1Loss()
+# optimizer = optim.Adam(model.parameters(), lr=0.001)
+# adamw 
+optimizer = optim.AdamW(model.parameters(), lr=0.001, weight_decay=0.001)
+#optimizer = optim.Adam(model.parameters(), lr=0.0005, weight_decay=0.0001)
+best_loss = np.inf  # Initialize the best loss to infinity
+start_epoch = 0
+scheduler = optim.lr_scheduler.StepLR(optimizer, step_size=10, gamma=0.5)
+#scheduler = optim.lr_scheduler.MultiStepLR(optimizer, milestones=[10, 20, 30], gamma=0.5)
+#scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer, mode='min', factor=0.5, patience=5, verbose=True)
+#scheduler = optim.lr_scheduler.CosineAnnealingLR(optimizer, T_max=10, eta_min=0.0001, last_epoch=-1)
+#scheduler = optim.lr_scheduler.OneCycleLR(optimizer, max_lr=0.001, steps_per_epoch=len(train_loader), epochs=epochs)
+
+load_saved = sys.argv[2] # "last, "best"
+if load_saved != "none":
+    with open(f"model_dumps/mix/{model_name}/{slice_count}/{sys.argv[1]}_best_model_with_metadata.pkl", "rb") as f:
+        checkpoint = pickle.load(f)
+    best_loss = checkpoint["loss"]
+
+    # Load the checkpoint
+    with open(f"model_dumps/mix/{model_name}/{slice_count}/{sys.argv[1]}_{load_saved}_model_with_metadata.pkl", "rb") as f:
+        checkpoint = pickle.load(f)
+
+    # Restore model and optimizer state
+    model.load_state_dict(checkpoint["model_state"])
+    optimizer.load_state_dict(checkpoint["optimizer_state"])
+
+    # Restore RNG states
+    torch.set_rng_state(checkpoint["t_rng_st"])
+    np.random.set_state(checkpoint["n_rng_st"])
+    if torch.cuda.is_available() and checkpoint["cuda_rng_st"] is not None:
+        torch.cuda.set_rng_state_all(checkpoint["cuda_rng_st"])
+
+    # Retrieve metadata
+    start_epoch = checkpoint["epoch"] + 1
+    loaded_loss = checkpoint["loss"]
+
+    print(f"Loaded model from epoch {start_epoch} with validation loss {loaded_loss:.4f}, best loss {best_loss:.4f}")
+
+
+predicted_ages = None
+# Training loop
+epochs = int(sys.argv[3])
+
+# Initialize lists to track loss
+filename = sys.argv[1] 
+csv_file = f"model_dumps/mix/{model_name}/{slice_count}/{filename}.csv"
+
+# Load existing epoch data if the file exists
+if os.path.exists(csv_file):
+    epoch_data = pd.read_csv(csv_file).to_dict(orient="records")
+    print(f"Loaded existing epoch data from {csv_file}.")
+else:
+    epoch_data = []
+    print("No existing epoch data found. Starting fresh.")
+
+
+# Plot loss vs. epoch and save the figure
+def update_loss_plot(epoch_data, filename):
+    df = pd.DataFrame(epoch_data)
+    df.to_csv(f"model_dumps/mix/{slice_count}/{filename}.csv", index=False)  # Save the data to CSV
+    
+    plt.figure(figsize=(8, 6))
+    plt.plot(df['epoch'], df['train_loss'], label="Train Loss", marker="o")
+    plt.plot(df['epoch'], df['val_loss'], label="Validation Loss", marker="o")
+    plt.xlabel("Epoch")
+    plt.ylabel("Loss")
+    plt.title("Loss vs. Epoch")
+    plt.legend()
+    plt.grid(True)
+    plt.savefig(f"model_dumps/mix/{slice_count}/{filename}.png")
+    plt.close()
+
+# Training loop
+for epoch in range(start_epoch, epochs):
+    model.train()
+    train_loss = 0.0
+    predicted_ages = []
+
+    for idx, (features, sex, age) in enumerate(train_loader):
+        features = features.unsqueeze(1).to(device)  # Add channel dimension
+        sex = sex.to(device)
+        age = age.to(device)
+        optimizer.zero_grad()
+        outputs = model(features, sex)
+        # Store the output for each sample in the batch
+        for i in range(outputs.size(0)):
+            train_outputs[train_indices[idx * batch_size + i]] = outputs[i].item()
+
+        loss = criterion(outputs.squeeze(), age)
+        loss.backward()
+        optimizer.step()
+
+        train_loss += loss.item()
+
+    train_loss /= len(train_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Train Loss: {train_loss:.4f}")
+
+    # Validation loop
+    model.eval()
+    val_loss = 0.0
+    with torch.no_grad():
+        for idx, (features, sex, age) in enumerate(val_loader):
+            features = features.unsqueeze(1).to(device)
+            sex = sex.to(device)
+            age = age.to(device)
+
+            outputs = model(features, sex)
+            loss = eval_crit(outputs.squeeze(), age)
+            val_loss += loss.item()
+
+            # Save the predicted age for the current validation sample
+            for i in range(outputs.size(0)):
+                val_outputs[val_indices[idx * batch_size + i]] = outputs[i].item()
+            # val_outputs[val_indices[idx]] = outputs.item()
+
+    val_loss /= len(val_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Validation Loss: {val_loss:.4f}")
+
+    #scheduler.step()
+
+    # Save the last model with metadata
+    print(f"Saving last model...")
+    checkpoint = {
+        "model_state": model.state_dict(),
+        "optimizer_state": optimizer.state_dict(),
+        "epoch": epoch,
+        "loss": val_loss,
+        "t_rng_st": torch.get_rng_state(),
+        "n_rng_st": np.random.get_state(),
+        "cuda_rng_st": torch.cuda.get_rng_state_all() if torch.cuda.is_available() else None
+    }
+    with open(f"model_dumps/mix/{model_name}/{slice_count}/{sys.argv[1]}_last_model_with_metadata.pkl", "wb") as f:
+        pickle.dump(checkpoint, f)
+    print(f"Last model saved...")
+
+    # Check if validation loss improved
+    if val_loss < best_loss:
+        best_loss = val_loss
+        print(f"Validation loss improved to {best_loss:.4f}. Saving model...")
+        with open(f"model_dumps/mix/{model_name}/{slice_count}/{sys.argv[1]}_best_model_with_metadata.pkl", "wb") as f:
+            pickle.dump(checkpoint, f)
+
+    # Save predictions and create DataFrames (same as before)
+    # ...
+
+    # Update epoch data and save the loss plot
+    epoch_data.append({
+        "epoch": epoch + 1,
+        "train_loss": train_loss,
+        "val_loss": val_loss
+    })
+    update_loss_plot(epoch_data, sys.argv[1])
+
+    max_index = max(train_outputs.keys())
+    # Create a DataFrame with NaN for all indices initially
+    df_trn = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
+    # Assign the values to their respective indices
+    for index, value in train_outputs.items():
+        df_trn.loc[index, "Predicted_Age"] = value
+    # print (df_trn)
+
+    df2 = df.copy()
+    df2['Predicted_Age'] = df_trn['Predicted_Age']
+    train_df = df2.loc[train_outputs.keys()]
+    # print (train_df)
+    train_df.to_csv(f"model_dumps/mix/{model_name}/{slice_count}/{sys.argv[1]}_predicted_ages_train.csv")
+
+    max_index = max(val_outputs.keys())
+    # Create a DataFrame with NaN for all indices initially
+    df_pred = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
+    # Assign the values to their respective indices
+    for index, value in val_outputs.items():
+        df_pred.loc[index, "Predicted_Age"] = value
+    # print (df_pred)
+
+    df1 = df.copy()
+    df1['Predicted_Age'] = df_pred['Predicted_Age']
+    test_df = df1.loc[val_outputs.keys()]
+    print (test_df)
+    test_df.to_csv(f"model_dumps/mix/{model_name}/{slice_count}/{sys.argv[1]}_predicted_ages_val.csv")
+
+
+    # Map unique first 4 characters of ImageID to color codes
+    unique_groups = test_df['ImageID'].str[:3].unique()
+    group_to_color = {group: i for i, group in enumerate(unique_groups)}
+
+    # Assign colors based on the mapping
+    cmap = plt.get_cmap('tab10')  # Change colormap as desired
+    colors = [cmap(group_to_color[group]) for group in test_df['ImageID'].str[:3]]
+
+    # Check that the predictions have been added to the DataFrame
+    # Plot Age vs. Predicted Age
+    plt.figure(figsize=(8, 6))
+    plt.scatter(test_df['Age'], test_df['Predicted_Age'], color=colors, label='Predicted vs Actual')
+    # plt.plot(test_df['Age'], test_df['Age'], color='red', linestyle='--', label='Perfect Prediction')  # Optional: Line of perfect prediction
+    # Add legend for colors based on ImageID groups
+    handles = [plt.Line2D([0], [0], marker='o', color=cmap(i), linestyle='', markersize=10) 
+            for i, group in enumerate(unique_groups)]
+    plt.legend(handles, unique_groups, title="ImageID Groups")
+    plt.xlabel('Age')
+    plt.ylabel('Predicted Age')
+    plt.title('Age vs Predicted Age')
+    plt.grid(True)
+    plt.savefig(f"model_dumps/mix/{model_name}/{slice_count}/plots/{model_name}_cnn_{sys.argv[1]}_epoch{epoch}.png")
diff --git a/vit_train_metrics.csv b/vit_train_metrics.csv
new file mode 100644
index 0000000..84a52db
--- /dev/null
+++ b/vit_train_metrics.csv
@@ -0,0 +1,3 @@
+3,1.153491705672757,57.71790919282511
+4,0.9176502034210103,66.72155269058295
+5,0.7490490353697569,73.3639293721973
diff --git a/vit_train_metrics_16.csv b/vit_train_metrics_16.csv
new file mode 100644
index 0000000..ed8230a
--- /dev/null
+++ b/vit_train_metrics_16.csv
@@ -0,0 +1,15 @@
+1,1.9349344514911753,30.69646860986547
+2,1.4820295305663695,45.65232623318386
+3,1.184266876237208,56.66690022421525
+4,0.9285762824072198,66.49383408071749
+5,0.7777999946643962,72.16227578475336
+1,1.9349344514911753,30.69646860986547
+2,1.4789821329639363,45.59627242152467
+3,1.187399784704428,56.07132847533632
+4,0.9425200620883316,65.6635369955157
+5,0.7542759179254585,72.84893497757848
+1,18.520315172827296
+2,17.078083339961655
+3,17.15616446051897
+4,17.168859794000873
+5,17.010499005226812
diff --git a/vit_train_metrics_32.csv b/vit_train_metrics_32.csv
new file mode 100644
index 0000000..963b9f0
--- /dev/null
+++ b/vit_train_metrics_32.csv
@@ -0,0 +1,38 @@
+1,17.599097248283737
+2,15.433740411070698
+3,12.161920863249644
+4,10.272737560919047
+5,9.215972374606828
+6,8.64863115676048
+7,8.578263686777765
+8,7.895704537496439
+9,7.318833008809475
+10,7.07903076324094
+11,6.877992988835536
+12,6.5025946226288385
+13,6.597799714637978
+14,6.411531273659836
+15,6.346511487034672
+1,17.60040367805637
+2,16.76178139160834
+3,15.88713075771861
+1,1.6977335079593976,38.03951793721973
+2,1.1888967110742001,55.68595852017937
+3,0.8706669232791349,67.72351457399103
+1,1.6995013216719232,37.90288677130045
+2,1.1868429208266227,55.79456278026906
+3,0.8673531520994789,67.80759529147981
+4,0.6384090962587446,76.73591647982063
+5,0.4965632288418387,82.23794843049328
+6,0.40968506034940855,85.81137892376681
+7,0.3472242637618842,88.16388733183857
+8,0.3045740294707966,89.69660874439462
+9,0.2790904786823196,90.50413396860986
+10,0.25063995638637576,91.61119674887892
+1,1.768086351878783,35.719412544169614
+2,1.286926638336635,52.13808524734982
+3,0.984793563152897,63.55316916961131
+4,0.7407006252659468,73.20837014134275
+5,0.5799919746269253,79.46113074204948
+1,1.7682361794543509,35.78290636042403
+6,0.47092240503395,83.49160777385158
diff --git a/vit_train_reg_feature_cnn_main_mix_roi.py b/vit_train_reg_feature_cnn_main_mix_roi.py
new file mode 100644
index 0000000..2c3fa71
--- /dev/null
+++ b/vit_train_reg_feature_cnn_main_mix_roi.py
@@ -0,0 +1,1286 @@
+import os
+import pandas as pd
+import nibabel as nib
+import numpy as np
+import torch
+import torch.nn as nn
+import torch.nn.init as init
+import torch.optim as optim
+from tqdm import tqdm
+from nibabel.orientations import io_orientation, axcodes2ornt, ornt_transform, apply_orientation
+from torchvision import transforms
+from transformers import ViTFeatureExtractor, ViTModel
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler
+import pickle
+import matplotlib.pyplot as plt
+import sys
+import SimpleITK as sitk
+from scipy.ndimage import zoom
+from dataset_cls import ADNIDataset, ADNIDatasetViT
+from torch.utils.data import DataLoader, Dataset
+import gc
+
+def set_random_seed(seed=69420):
+    np.random.seed(seed)
+    torch.manual_seed(seed)
+    torch.cuda.manual_seed(seed)
+    torch.cuda.manual_seed_all(seed)  # For multi-GPU setups
+    torch.backends.cudnn.deterministic = True
+    torch.backends.cudnn.benchmark = False
+
+universal_seed = 69420
+
+set_random_seed(universal_seed)
+
+# Function to resample the volume to 160 slices
+def resample_volume_to_fixed_slices(data, affine, target_slices=160):
+    # Convert Numpy array and affine to SimpleITK image
+    sitk_img = sitk.GetImageFromArray(data)
+    sitk_img.SetSpacing([affine[0, 0], affine[1, 1], affine[2, 2]])
+
+    original_size = sitk_img.GetSize()  # (width, height, depth)
+    original_spacing = sitk_img.GetSpacing()  # (spacing_x, spacing_y, spacing_z)
+
+    # Calculate new spacing to achieve the target number of slices
+    new_spacing = list(original_spacing)
+    new_spacing[2] = (original_spacing[2] * original_size[2]) / target_slices
+
+    # Define new size
+    new_size = [original_size[0], original_size[1], target_slices]
+
+    # Resample the image
+    resampler = sitk.ResampleImageFilter()
+    resampler.SetOutputSpacing(new_spacing)
+    resampler.SetSize(new_size)
+    resampler.SetInterpolator(sitk.sitkLinear)
+    resampled_img = resampler.Execute(sitk_img)
+
+    return sitk.GetArrayFromImage(resampled_img)  # Return the resampled image as a numpy array
+
+
+def resample_nifti(img_data, target_slices = 160):
+    # Determine the current number of slices along the z-axis (3rd dimension)
+    current_slices = img_data.shape[0]
+    # Calculate the zoom factor for resampling (only along the z-axis)
+    zoom_factor = target_slices / current_slices
+    # Resample the image data along the z-axis
+    resampled_data = zoom(img_data, (zoom_factor, 1, 1), order=3)  # order=3 for cubic interpolation
+    # Ensure that the resampled data has the target number of slices
+    # print (resampled_data.shape)
+    # resampled_data = resampled_data[:target_slices,:,:]
+    # print (resampled_data.shape)
+    return resampled_data
+
+
+# Check if GPU is available
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+# Load the CSV file into a pandas DataFrame
+csv_path = "adni_storage/adni_brainrotnet_metadata.csv"
+df_adni = pd.read_csv(csv_path)
+# df = df.sample(n=1000, random_state=69420)
+# Add a new column 'filepath' with the constructed file paths
+df_adni['filepath'] = df_adni.apply(
+    lambda row: f"adni_storage/ADNI_nii_gz_bias_corrected/I{row['ImageID'][4:]}_{row['SubjectID']}.stripped.N4.nii.gz",
+    axis=1
+)
+df_adni = df_adni.sort_values(by='Age', ascending=True).reset_index(drop=True).head(500)
+# df_adni=df_adni.sample(n=400)
+
+# Load independent dataset metadata
+# metadata_path = "ixi_storage/ixi_brainrotnet_metadata.csv"
+# df_ixi = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_ixi['filepath'] = df_ixi.apply(
+#     lambda row: f"ixi_storage/IXI_bias_corrected/{row['ImageID']}.stripped.N4.nii.gz",
+#     axis=1
+# )
+
+# metadata_path = "abide_storage/abide_brainrotnet_metadata.csv"
+# df_abide = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_abide['filepath'] = df_abide.apply(
+#     lambda row: f"abide_storage/ABIDEII_bias_corrected/{row['ImageID'][7:]}.stripped.N4.nii.gz",
+#     axis=1
+# )
+# df_abide = df_abide.sort_values(by='Age', ascending=False).reset_index(drop=True).head(500)
+# df_abide=df_abide.sample(n=400)
+
+metadata_path = "dlbs_storage/dlbs_brainrotnet_metadata.csv"
+df_dlbs = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_dlbs['filepath'] = df_dlbs.apply(
+    lambda row: f"dlbs_storage/DLBS_bias_corrected/{row['ImageID'][4:]}.stripped.N4.nii.gz",
+    axis=1
+)
+
+# metadata_path = "cobre_storage/cobre_brainrotnet_metadata.csv"
+# df_cobre = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_cobre['filepath'] = df_cobre.apply(
+#     lambda row: f"cobre_storage/COBRE_bias_corrected/{row['ImageID'][5:]}.stripped.N4.nii.gz",
+#     axis=1
+# )
+
+metadata_path = "fcon1000_storage/fcon1000_brainrotnet_metadata.csv"
+df_fcon = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_fcon['filepath'] = df_fcon.apply(
+    lambda row: f"fcon1000_storage/fcon1000_bias_corrected/{row['ImageID'][8:]}.stripped.N4.nii.gz",
+    axis=1
+)
+df_fcon = df_fcon.dropna()
+# df_fcon = df_fcon.sort_values(by='Age', ascending=False).reset_index(drop=True).head(300)
+# df_fcon = df_fcon.sample(n=300)
+
+# metadata_path = "sald_storage/sald_brainrotnet_metadata.csv"
+# df_sald = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_sald['filepath'] = df_sald.apply(
+#     lambda row: f"sald_storage/SALD_bias_corrected/sub-{row['ImageID'][4:]}.stripped.N4.nii.gz",
+#     axis=1
+# )
+# df_sald = df_sald.sort_values(by='Age', ascending=False).reset_index(drop=True).head(300)
+# # df_sald = df_sald.sample(n=300)
+
+# metadata_path = "corr_storage/corr_brainrotnet_metadata.csv"
+# df_corr = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_corr['filepath'] = df_corr.apply(
+#     lambda row: f"corr_storage/CORR_bias_corrected/{row['ImageID'][5:]}.stripped.N4.nii.gz",
+#     axis=1
+# )
+# df_corr = df_corr.sort_values(by='Age', ascending=True).reset_index(drop=True).head(300)
+# df_corr = df_corr.sample(n=200)
+
+
+# metadata_path = "oasis1_storage/oasis1_brainrotnet_metadata.csv"
+# df_oas1 = pd.read_csv(metadata_path)
+# # Update filepaths for the independent dataset
+# df_oas1['filepath'] = df_oas1.apply(
+#     lambda row: f"oasis1_storage/oasis_nii_gz_bias_corrected/{row['ImageID']}.stripped.N4.nii.gz",
+#     axis=1
+# )
+# df_oas1 = df_oas1.sort_values(by='Age', ascending=False).reset_index(drop=True).head(300)
+metadata_path = "camcan_storage/camcan_brainrotnet_metadata.csv"
+df_camcan = pd.read_csv(metadata_path)
+# Update filepaths for the independent dataset
+df_camcan['filepath'] = df_camcan.apply(
+    lambda row: f"camcan_storage/CamCAN_nii_gz_bias_corrected/{row['ImageID']}.stripped.N4.nii.gz",
+    axis=1
+)
+
+
+df = pd.concat ([
+                 df_adni[['ImageID', 'Sex', 'Age', 'filepath']], 
+                #  df_ixi[['ImageID', 'Sex', 'Age', 'filepath']], 
+                #  df_abide[['ImageID', 'Sex', 'Age', 'filepath']],
+                 df_dlbs[['ImageID', 'Sex', 'Age', 'filepath']],
+                #  df_cobre[['ImageID', 'Sex', 'Age', 'filepath']],
+                 df_fcon[['ImageID', 'Sex', 'Age', 'filepath']],
+                #  df_sald[['ImageID', 'Sex', 'Age', 'filepath']],
+                #  df_corr[['ImageID', 'Sex', 'Age', 'filepath']], 
+                #  df_oas1[['ImageID', 'Sex', 'Age', 'filepath']],
+                df_camcan[['ImageID', 'Sex', 'Age', 'filepath']]
+                 ], ignore_index=True)
+print (df)
+# Ensure 'Age' is an integer
+# df['Age_Group'] = df['Age'].astype(int).apply(lambda x: f"{x:03d}"[:-1] + "0")
+# df['Age_Group'] = df['Age_Group'] + df['Sex']
+# print (df['Age_Group'].unique())
+# Prepare dataset and dataloaders
+sex_encoded = df['Sex'].apply(lambda x: 0 if x == 'M' else 1).tolist()
+age_list = df['Age'].tolist()
+filepath_list = df['filepath'].tolist()
+# label_list = df['Age_Group'].tolist()
+# Use 'Age' directly as labels
+label_list = df['Age'].tolist()
+
+# Get unique labels and create a mapping
+# unique_labels = sorted(set(label_list))  # Ensure consistent ordering
+# label_to_idx = {label: idx for idx, label in enumerate(unique_labels)}
+# idx_to_label = {idx: label for label, idx in label_to_idx.items()}  # Reverse mapping for decoding
+
+# # Convert labels to integers
+# numeric_labels = [label_to_idx[label] for label in label_list]
+# label_list = numeric_labels
+
+roi = 160
+
+# Transformation pipeline for ViT
+transform = transforms.Compose([
+    transforms.ToPILImage(),
+    transforms.Resize((224, 224)),
+    transforms.Lambda(lambda img: img.convert("RGB")),  # Convert to RGB
+    transforms.ToTensor(),
+    transforms.Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5]),  # Normalize for ViT
+])
+
+
+# Function to extract 16 evenly spaced slices
+def extract_slices(volume, num_slices=16):
+    total_slices = volume.shape[0]
+    indices = np.linspace(0, total_slices - 1, num_slices, dtype=int)
+    return volume[indices, :, :]  # Select slices
+
+def calculate_bounding_box_from_volume(volume):
+    # Find indices of non-zero values
+    non_zero_indices = np.argwhere(volume > 0)
+
+    # Calculate min and max indices along each dimension
+    min_indices = np.min(non_zero_indices, axis=0)
+    max_indices = np.max(non_zero_indices, axis=0)
+
+    # Convert indices to integers
+    min_indices = min_indices.astype(int)
+    max_indices = max_indices.astype(int)
+
+    return min_indices, max_indices
+
+
+def crop_brain_volumes(brain_data):
+    
+
+        # Calculate bounding box from the brain volume
+    min_indices, max_indices = calculate_bounding_box_from_volume(brain_data)
+
+        # Crop the volume
+    cropped_brain = brain_data[min_indices[0]:max_indices[0] + 1,
+                                   min_indices[1]:max_indices[1] + 1,
+                                   min_indices[2]:max_indices[2] + 1]
+    return cropped_brain
+
+# Function to preprocess data and dynamically expand slices while saving to disk
+def preprocess_and_expand(dataset, transform, output_dir, num_slices=16):
+    os.makedirs(output_dir, exist_ok=True)  # Ensure output directory exists
+    expanded_images, expanded_labels = [], []
+
+    for filepath, label in tqdm(dataset, desc="Processing Slices"):
+        # Check if all slice files already exist
+        all_slices_exist = True
+        slice_filenames = [
+            os.path.join(output_dir, f"{os.path.basename(filepath)}_slice_{i}.pt")
+            for i in range(num_slices)
+        ]
+        if not all(os.path.exists(slice_file) for slice_file in slice_filenames):
+            all_slices_exist = False
+
+        # Skip processing if all slices exist
+        if all_slices_exist:
+            expanded_images.extend(slice_filenames)  # Add existing file paths
+            expanded_labels.extend([label] * num_slices)
+            continue
+
+        # Load NIfTI image only if slices are missing
+        nii_img = nib.load(filepath)
+        orig_ornt = io_orientation(nii_img.affine)
+        ras_ornt = axcodes2ornt(("R", "A", "S"))
+        ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+        data = nii_img.get_fdata()  # Load image data
+        data = apply_orientation(data, ornt_trans)
+
+        data = crop_brain_volumes(data)
+
+        # Normalize and extract slices
+        data = (data - data.min()) / (data.max() - data.min())
+        slices = extract_slices(data, num_slices)
+
+        # Transform each slice, save to file, and add to dataset
+        for i, slice_data in enumerate(slices):
+            slice_filename = slice_filenames[i]
+            if not os.path.exists(slice_filename):
+                transformed_slice = transform(slice_data)  # Transform slice
+                torch.save(transformed_slice, slice_filename)  # Save to file
+            expanded_images.append(slice_filename)  # Store file path
+            expanded_labels.append(label)
+
+    return expanded_images, expanded_labels
+# Instantiate Dataset
+vit_dataset = ADNIDatasetViT(filepath_list, label_list)
+
+# Split Dataset
+train_size = int(0.8 * len(vit_dataset))
+val_size = len(vit_dataset) - train_size
+generator = torch.Generator().manual_seed(universal_seed)
+vit_train_dataset, vit_val_dataset = torch.utils.data.random_split(vit_dataset, [train_size, val_size], generator=generator)
+
+# Create New Dataset with Filepaths
+class ExpandedDataset(Dataset):
+    def __init__(self, image_paths, labels):
+        self.image_paths = image_paths
+        self.labels = labels
+
+    def __len__(self):
+        return len(self.labels)
+
+    def __getitem__(self, idx):
+        # Load the image from file
+        image = torch.load(self.image_paths[idx])
+        label = torch.tensor(self.labels[idx], dtype=torch.long)
+        return image, label
+
+slice_count = 32
+
+# Define output directory for slices
+output_dir = f"processed_slices/{slice_count}"
+
+# Preprocess and expand the training data
+expanded_image_paths, expanded_labels = preprocess_and_expand(vit_train_dataset, transform, output_dir, num_slices=slice_count)
+
+# Create Expanded Dataset and DataLoader
+expanded_train_dataset = ExpandedDataset(expanded_image_paths, expanded_labels)
+expanded_train_loader = DataLoader(expanded_train_dataset, batch_size=8, shuffle=True)
+
+# Print Sizes
+print(f"Original Training Dataset Size: {len(vit_train_dataset)}")
+print(f"Expanded Training Dataset Size: {len(expanded_train_dataset)}")
+
+from transformers import ViTForImageClassification
+# Load ViT model
+# num_classes = df['Age_Group'].nunique()  # Number of unique Age_Groups
+model = ViTForImageClassification.from_pretrained(
+    "google/vit-base-patch16-224",
+    num_labels=1,  # Regression task
+    ignore_mismatched_sizes=True, 
+)
+
+model.to(device)
+vit_train_epochs = 3
+# Loss function and optimizer
+criterion = nn.L1Loss()
+optimizer = optim.AdamW(model.parameters(), lr=1e-4)
+
+# Function to save checkpoint
+def save_checkpoint(epoch, model, optimizer, path=f"model_dumps/vit_train_reg_checkpoint_slice{slice_count}_e{vit_train_epochs}.pth"):
+    torch.save({
+        'epoch': epoch,
+        'model_state_dict': model.state_dict(),
+        'optimizer_state_dict': optimizer.state_dict(),
+    }, path)
+    print(f"Checkpoint saved at epoch {epoch+1}")
+
+# Function to load checkpoint
+def load_checkpoint(path=f"model_dumps/vit_train_reg_checkpoint_slice{slice_count}_e{vit_train_epochs}.pth"):
+    checkpoint = torch.load(path)
+    model.load_state_dict(checkpoint['model_state_dict'])
+    optimizer.load_state_dict(checkpoint['optimizer_state_dict'])
+    start_epoch = checkpoint['epoch'] + 1
+    print(f"Checkpoint loaded. Resuming from epoch {start_epoch}")
+    return start_epoch
+
+# Check if recovery mode is enabled
+checkpoint_path = f"model_dumps/vit_train_reg_checkpoint_slice{slice_count}_e{vit_train_epochs}.pth"
+start_epoch = 0
+
+if len(sys.argv) > 4 and sys.argv[4] == "recover":
+    start_epoch = load_checkpoint(path=checkpoint_path)
+
+# Training loop
+
+model.train()
+
+for epoch in range(start_epoch, vit_train_epochs):
+    running_loss = 0.0
+    correct = 0
+    total_samples = 0
+
+    for inputs, labels in tqdm(expanded_train_loader, desc=f"Epoch {epoch+1}/{vit_train_epochs}"):
+        # Move data to device
+        inputs, labels = inputs.to(device), labels.to(device)
+
+        # Forward pass
+        outputs = model(pixel_values=inputs)  # ViT expects `pixel_values`
+        predictions = outputs.logits.squeeze()  # Squeeze logits for regression (shape: [batch_size])
+
+        # Calculate loss
+        loss = criterion(predictions, labels)
+
+        # Backward pass and optimization
+        optimizer.zero_grad()
+        loss.backward()
+        optimizer.step()
+
+        running_loss += loss.item() * labels.size(0)  # Sum up the loss
+        total_samples += labels.size(0)
+
+    # Calculate epoch metrics
+    epoch_mae = running_loss / total_samples  # Loss is MAE in this case
+    # epoch_mae = running_loss / len(expanded_train_loader)
+    print(f"Epoch {epoch+1}/{vit_train_epochs}, MAE (Loss): {epoch_mae:.4f}")
+    print(f"Total Samples: {total_samples}")
+
+    # Save the accuracy and loss for each epoch in a CSV file
+    with open(f'vit_train_metrics_{slice_count}.csv', 'a') as f:
+        f.write(f"{epoch+1},{epoch_mae}\n")
+   
+    save_checkpoint(epoch, model, optimizer, path=f"model_dumps/vit_train_reg_checkpoint_slice{slice_count}_e{vit_train_epochs}.pth")
+    gc.collect()
+
+# Load pre-trained ViT model
+feature_extractor = ViTFeatureExtractor.from_pretrained("google/vit-base-patch16-224")
+# model = ViTModel.from_pretrained("google/vit-base-patch16-224")
+# model.to(device)  # Move the model to the GPU (if available)
+model.eval()
+
+# Update image transform for grayscale images to match ViT input requirements
+transform = transforms.Compose([
+    transforms.ToPILImage(),
+    transforms.Resize((224, 224)),
+    transforms.Lambda(lambda img: img.convert("RGB")),  # Convert to RGB directly
+    transforms.ToTensor(),
+    transforms.Normalize(mean=feature_extractor.image_mean, std=feature_extractor.image_std),
+])
+
+torch.cuda.empty_cache()  # Free GPU memory
+
+
+
+# To store features and labels
+features_list = []
+labels_list = []
+
+
+# Directory to save processed images and features
+os.makedirs(f"adni_storage/ADNI_features/train_e{vit_train_epochs}_reg/{slice_count}/", exist_ok=True)
+# Process each row in the DataFrame
+for _, row in tqdm(df_adni.iterrows(), total=len(df_adni), desc="Processing images"):
+    filepath = row['filepath']
+    image_title = f"{row['ImageID'][4:]}_{row['SubjectID']}"
+
+    # Check if the feature file already exists
+    feature_file_path = f"adni_storage/ADNI_features/train_e{vit_train_epochs}_reg/{slice_count}/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        
+        from PIL import Image
+        # Normalize the array to 0-255 for grayscale image
+        data_normalized = ((features - np.min(features)) / (np.max(features) - np.min(features)) * 255).astype(np.uint8)
+        data_normalized = np.repeat(data_normalized, 4, axis=0)
+        # Create an image from the array
+        img = Image.fromarray(np.transpose(data_normalized), mode='L')  # 'L' mode for grayscale
+        # Save the image
+        # img.save(f"adni_storage/ADNI_features/train_e{vit_train_epochs}_reg/{slice_count}/featuremaps/{image_title}_fm.png")
+
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        # print ("hiii")
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                data = crop_brain_volumes(data)
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+
+                # Extract features for all sagittal slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+
+                    # Transform slice for ViT input
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)  # Add batch dimension and move to GPU
+
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        # #outputs = model(slice_tensor)
+                        # slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()  # Move output back to CPU
+                        slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Target is 'Age'
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+
+# Directory to save processed images and features
+# os.makedirs(f"ixi_storage/IXI_features/train_e{vit_train_epochs}_reg/{slice_count}/", exist_ok=True)
+# for _, row in tqdm(df_ixi.iterrows(), total=len(df_ixi), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID']}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}_reg/{slice_count}/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         #outputs = model(slice_tensor)
+#                         slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+
+# # Directory to save processed images and features
+# os.makedirs(f"abide_storage/ABIDEII_features/train_e{vit_train_epochs}_reg/{slice_count}/", exist_ok=True)
+# for _, row in tqdm(df_abide.iterrows(), total=len(df_abide), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][7:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"abide_storage/ABIDEII_features/train_e{vit_train_epochs}_reg/{slice_count}/{image_title}_features.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}_reg/{slice_count}/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         #outputs = model(slice_tensor)
+#                         slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+os.makedirs(f"dlbs_storage/DLBS_features/train_e{vit_train_epochs}_reg/{slice_count}/", exist_ok=True)
+for _, row in tqdm(df_dlbs.iterrows(), total=len(df_dlbs), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID'][4:]}"
+        # Check if the feature file already exists
+    feature_file_path = f"dlbs_storage/DLBS_features/train_e{vit_train_epochs}_reg/{slice_count}/{image_title}_features.npy"
+    # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}_reg/{slice_count}/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                data = crop_brain_volumes(data)
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        #outputs = model(slice_tensor)
+                        slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+
+# os.makedirs(f"cobre_storage/COBRE_features/train_e{vit_train_epochs}_reg/{slice_count}/", exist_ok=True)
+# for _, row in tqdm(df_cobre.iterrows(), total=len(df_cobre), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][5:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"cobre_storage/COBRE_features/train_e{vit_train_epochs}_reg/{slice_count}/{image_title}_features.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}_reg/{slice_count}/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         #outputs = model(slice_tensor)
+#                         slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+os.makedirs(f"fcon1000_storage/fcon1000_features/train_e{vit_train_epochs}_reg/{slice_count}/", exist_ok=True)
+for _, row in tqdm(df_fcon.iterrows(), total=len(df_fcon), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID'][5:]}"
+        # Check if the feature file already exists
+    feature_file_path = f"fcon1000_storage/fcon1000_features/train_e{vit_train_epochs}_reg/{slice_count}/{image_title}_features.npy"
+    # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}_reg/{slice_count}/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                data = crop_brain_volumes(data)
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        #outputs = model(slice_tensor)
+                        slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+# os.makedirs(f"sald_storage/SALD_features/train_e{vit_train_epochs}_reg/{slice_count}/", exist_ok=True)
+# for _, row in tqdm(df_sald.iterrows(), total=len(df_sald), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][4:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"sald_storage/SALD_features/train_e{vit_train_epochs}_reg/{slice_count}/{image_title}_features.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}_reg/{slice_count}/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         #outputs = model(slice_tensor)
+#                         slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+# os.makedirs(f"corr_storage/CORR_features/train_e{vit_train_epochs}_reg/{slice_count}/", exist_ok=True)
+# for _, row in tqdm(df_corr.iterrows(), total=len(df_corr), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID'][5:]}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"corr_storage/CORR_features/train_e{vit_train_epochs}_reg/{slice_count}/{image_title}_features.npy"
+#     # feature_file_path = f"ixi_storage/IXI_features/train_e{vit_train_epochs}_reg/{slice_count}/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         #outputs = model(slice_tensor)
+#                         slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+
+# os.makedirs(f"oasis1_storage/oasis1_features/train_e{vit_train_epochs}_reg/{slice_count}/", exist_ok=True)
+# for _, row in tqdm(df_oas1.iterrows(), total=len(df_oas1), desc="Processing test images"):
+#     filepath = row['filepath']    
+#     image_title = f"{row['ImageID']}"
+#         # Check if the feature file already exists
+#     feature_file_path = f"oasis1_storage/oasis1_features/train_e{vit_train_epochs}_reg/{slice_count}/{image_title}_features.npy"
+#     if os.path.exists(feature_file_path):
+#         # If file exists, load the features from the file
+#         features = np.load(feature_file_path)
+        
+#         features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+#         features_list.append(features)  # Flatten the features and add to the list
+#         labels_list.append(row['Age'])  # Add the corresponding age label
+#     else:
+#         if os.path.exists(filepath):
+#             try:
+#                 # Load the NIfTI image
+#                 nii_img = nib.load(filepath)
+
+#                 # Get current orientation and reorient to RAS
+#                 orig_ornt = io_orientation(nii_img.affine)
+#                 ras_ornt = axcodes2ornt(("R", "A", "S"))
+#                 ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+#                 data = nii_img.get_fdata()  # Load image data
+#                 data = apply_orientation(data, ornt_trans)
+
+#                 affine = nii_img.affine  # Affine transformation matrix
+
+#                 # Resample the volume to 160 slices (if required)
+#                 data = resample_nifti(data, target_slices=160)
+#                 # Extract features for all slices
+#                 features = []
+#                 for slice_idx in range(data.shape[0]):
+#                     slice_data = data[slice_idx, :, :]
+#                     slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+#                     slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+#                     # Extract features using ViT
+#                     with torch.no_grad():
+#                         #outputs = model(slice_tensor)
+#                         slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+#                         features.append(slice_features)
+#                 # Save extracted features
+#                 features = np.array(features)
+#                 np.save(feature_file_path, features)
+#                 features_list.append(features)
+#                 labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+#             except Exception as e:
+#                 print(f"Error processing {filepath}: {e}")
+#         else:
+#             print(f"File not found: {filepath}")
+
+os.makedirs(f"camcan_storage/CamCAN_features/train_e{vit_train_epochs}_reg/{slice_count}/", exist_ok=True)
+for _, row in tqdm(df_camcan.iterrows(), total=len(df_camcan), desc="Processing test images"):
+    filepath = row['filepath']    
+    image_title = f"{row['ImageID']}"
+        # Check if the feature file already exists
+    feature_file_path = f"camcan_storage/CamCAN_features/train_e{vit_train_epochs}_reg/{slice_count}/{image_title}_features.npy"
+    # feature_file_path = f"ixi_storage/IXI_features/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        
+        features =  features[len(features) // 2 - roi//2 : len(features) // 2 + roi//2]
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                data = crop_brain_volumes(data)
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_nifti(data, target_slices=160)
+                # Extract features for all slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+                    
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)
+                    
+                    # Extract features using ViT
+                    with torch.no_grad():
+                      
+                        slice_features = model.vit(slice_tensor).last_hidden_state.mean(dim=1).squeeze().cpu().numpy()
+                        features.append(slice_features)
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Assuming 'Age' is the target
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+
+
+batch_size = 1
+
+# print (features_list)
+print (features_list[0].shape)
+
+# Create Dataset and DataLoader
+dataset = ADNIDataset(features_list, sex_encoded, age_list)
+train_size = int(0.8 * len(dataset))
+val_size = len(dataset) - train_size
+generator.manual_seed(universal_seed)
+train_dataset, val_dataset = torch.utils.data.random_split(dataset, [train_size, val_size], generator=generator)
+# Store the indices of the validation dataset
+val_indices = val_dataset.indices
+train_indices = train_dataset.indices
+
+train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=False)
+val_loader = DataLoader(val_dataset, batch_size=batch_size, shuffle=False)
+
+# Tracking outputs for validation samples
+val_outputs = {}
+train_outputs = {}
+
+# Initialize model, loss, and optimizer
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+
+###########################################
+# THIS IS WHERE YOU CHOOSE A MODEL TO TEST 
+###########################################
+
+import importlib
+
+# Assuming sys.argv[1] is the module name
+module_name = sys.argv[1]  # Example: "my_model"
+class_name = "AgePredictionCNN"  # The class you want to import
+
+try:
+    # Dynamically import the module
+    module = importlib.import_module(module_name)
+    
+    # Dynamically get the class
+    AgePredictionCNN = getattr(module, class_name)
+    
+    print(f"Successfully imported {class_name} from {module_name}.")
+
+except ImportError:
+    print(f"Module {module_name} could not be imported.")
+except AttributeError:
+    print(f"{class_name} does not exist in {module_name}.")
+
+##############################
+# MODEL IMPORTED DYNAMICALLY
+##############################
+
+print (features_list[0].shape)
+model = AgePredictionCNN((1, features_list[0].shape[0], features_list[0].shape[1])).to(device)
+criterion = nn.MSELoss()
+
+eval_crit = nn.L1Loss()
+optimizer = optim.Adam(model.parameters(), lr=0.001)
+best_loss = np.inf  # Initialize the best loss to infinity
+start_epoch = 0
+
+
+load_saved = sys.argv[2] # "last, "best"
+if load_saved != "none":
+    with open(f"model_dumps/mix/{slice_count}/reg_{sys.argv[1]}_best_model_with_metadata_{vit_train_epochs}.pkl", "rb") as f:
+        checkpoint = pickle.load(f)
+    best_loss = checkpoint["loss"]
+
+    # Load the checkpoint
+    with open(f"model_dumps/mix/{slice_count}/reg_{sys.argv[1]}_{load_saved}_model_with_metadata_{vit_train_epochs}.pkl", "rb") as f:
+        checkpoint = pickle.load(f)
+
+    # Restore model and optimizer state
+    model.load_state_dict(checkpoint["model_state"])
+    optimizer.load_state_dict(checkpoint["optimizer_state"])
+
+    # Restore RNG states
+    torch.set_rng_state(checkpoint["t_rng_st"])
+    np.random.set_state(checkpoint["n_rng_st"])
+    if torch.cuda.is_available() and checkpoint["cuda_rng_st"] is not None:
+        torch.cuda.set_rng_state_all(checkpoint["cuda_rng_st"])
+
+    # Retrieve metadata
+    start_epoch = checkpoint["epoch"] + 1
+    loaded_loss = checkpoint["loss"]
+
+    print(f"Loaded model from epoch {start_epoch} with validation loss {loaded_loss:.4f}, best loss {best_loss:.4f}")
+
+predicted_ages = None
+# Training loop
+epochs = int(sys.argv[3])
+
+# Initialize lists to track loss
+filename = sys.argv[1] 
+csv_file = f"model_dumps/mix/{slice_count}/{filename}_reg.csv"
+
+# Load existing epoch data if the file exists
+if os.path.exists(csv_file):
+    epoch_data = pd.read_csv(csv_file).to_dict(orient="records")
+    print(f"Loaded existing epoch data from {csv_file}.")
+else:
+    epoch_data = []
+    print("No existing epoch data found. Starting fresh.")
+
+
+# Plot loss vs. epoch and save the figure
+def update_loss_plot(epoch_data, filename):
+    df = pd.DataFrame(epoch_data)
+    df.to_csv(f"model_dumps/mix/{slice_count}/{filename}_reg.csv", index=False)  # Save the data to CSV
+    
+    plt.figure(figsize=(8, 6))
+    plt.plot(df['epoch'], df['train_loss'], label="Train Loss", marker="o")
+    plt.plot(df['epoch'], df['val_loss'], label="Validation Loss", marker="o")
+    plt.xlabel("Epoch")
+    plt.ylabel("Loss")
+    plt.title("Loss vs. Epoch")
+    plt.legend()
+    plt.grid(True)
+    plt.savefig(f"model_dumps/mix/{slice_count}/{filename}_reg.png")
+    plt.close()
+
+# Training loop
+for epoch in range(start_epoch, epochs):
+    model.train()
+    train_loss = 0.0
+    predicted_ages = []
+
+    for idx, (features, sex, age) in enumerate(train_loader):
+        features = features.unsqueeze(1).to(device)  # Add channel dimension
+        sex = sex.to(device)
+        age = age.to(device)
+        optimizer.zero_grad()
+        outputs = model(features, sex)
+        # Store the output for each sample in the batch
+        for i in range(outputs.size(0)):
+            train_outputs[train_indices[idx * batch_size + i]] = outputs[i].item()
+
+        loss = criterion(outputs.squeeze(), age)
+        loss.backward()
+        optimizer.step()
+
+        train_loss += loss.item()
+
+    train_loss /= len(train_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Train Loss: {train_loss:.4f}")
+
+    # Validation loop
+    model.eval()
+    val_loss = 0.0
+    with torch.no_grad():
+        for idx, (features, sex, age) in enumerate(val_loader):
+            features = features.unsqueeze(1).to(device)
+            sex = sex.to(device)
+            age = age.to(device)
+
+            outputs = model(features, sex)
+            loss = eval_crit(outputs.squeeze(), age)
+            val_loss += loss.item()
+
+            # Save the predicted age for the current validation sample
+            for i in range(outputs.size(0)):
+                val_outputs[val_indices[idx * batch_size + i]] = outputs[i].item()
+            # val_outputs[val_indices[idx]] = outputs.item()
+
+    val_loss /= len(val_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Validation Loss: {val_loss:.4f}")
+
+    # Save the last model with metadata
+    print(f"Saving last model...")
+    checkpoint = {
+        "model_state": model.state_dict(),
+        "optimizer_state": optimizer.state_dict(),
+        "epoch": epoch,
+        "loss": val_loss,
+        "t_rng_st": torch.get_rng_state(),
+        "n_rng_st": np.random.get_state(),
+        "cuda_rng_st": torch.cuda.get_rng_state_all() if torch.cuda.is_available() else None
+    }
+    with open(f"model_dumps/mix/{slice_count}/reg_{sys.argv[1]}_last_model_with_metadata_{vit_train_epochs}.pkl", "wb") as f:
+        pickle.dump(checkpoint, f)
+    print(f"Last model saved...")
+
+    # Check if validation loss improved
+    if val_loss < best_loss:
+        best_loss = val_loss
+        print(f"Validation loss improved to {best_loss:.4f}. Saving model...")
+        with open(f"model_dumps/mix/{slice_count}/reg_{sys.argv[1]}_best_model_with_metadata_{vit_train_epochs}.pkl", "wb") as f:
+            pickle.dump(checkpoint, f)
+
+    # Save predictions and create DataFrames (same as before)
+    # ...
+
+    # Update epoch data and save the loss plot
+    epoch_data.append({
+        "epoch": epoch + 1,
+        "train_loss": train_loss,
+        "val_loss": val_loss
+    })
+    update_loss_plot(epoch_data, sys.argv[1])
+
+    max_index = max(train_outputs.keys())
+    # Create a DataFrame with NaN for all indices initially
+    df_trn = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
+    # Assign the values to their respective indices
+    for index, value in train_outputs.items():
+        df_trn.loc[index, "Predicted_Age"] = value
+    # print (df_trn)
+
+    df2 = df.copy()
+    df2['Predicted_Age'] = df_trn['Predicted_Age']
+    train_df = df2.loc[train_outputs.keys()]
+    # print (train_df)
+    train_df.to_csv(f"model_dumps/mix/{slice_count}/reg_{sys.argv[1]}_predicted_ages_train.csv")
+
+    max_index = max(val_outputs.keys())
+    # Create a DataFrame with NaN for all indices initially
+    df_pred = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
+    # Assign the values to their respective indices
+    for index, value in val_outputs.items():
+        df_pred.loc[index, "Predicted_Age"] = value
+    # print (df_pred)
+
+    df1 = df.copy()
+    df1['Predicted_Age'] = df_pred['Predicted_Age']
+    test_df = df1.loc[val_outputs.keys()]
+    print (test_df)
+    test_df.to_csv(f"model_dumps/mix/{slice_count}/reg_{sys.argv[1]}_predicted_ages_val.csv")
+
+
+    # Map unique first 4 characters of ImageID to color codes
+    unique_groups = test_df['ImageID'].str[:3].unique()
+    group_to_color = {group: i for i, group in enumerate(unique_groups)}
+
+    # Assign colors based on the mapping
+    cmap = plt.get_cmap('tab10')  # Change colormap as desired
+    colors = [cmap(group_to_color[group]) for group in test_df['ImageID'].str[:3]]
+
+    # Check that the predictions have been added to the DataFrame
+    # Plot Age vs. Predicted Age
+    plt.figure(figsize=(8, 6))
+    plt.scatter(test_df['Age'], test_df['Predicted_Age'], color=colors, label='Predicted vs Actual')
+    # plt.plot(test_df['Age'], test_df['Age'], color='red', linestyle='--', label='Perfect Prediction')  # Optional: Line of perfect prediction
+    # Add legend for colors based on ImageID groups
+    handles = [plt.Line2D([0], [0], marker='o', color=cmap(i), linestyle='', markersize=10) 
+            for i, group in enumerate(unique_groups)]
+    plt.legend(handles, unique_groups, title="ImageID Groups")
+    plt.xlabel('Age')
+    plt.ylabel('Predicted Age')
+    plt.title('Age vs Predicted Age')
+    plt.grid(True)
+    plt.savefig(f"model_dumps/mix/{slice_count}/plots/reg_vit_cnn_{sys.argv[1]}_epoch{epoch}.png")
diff --git a/vitlarge_feature_cnn_main.py b/vitlarge_feature_cnn_main.py
new file mode 100644
index 0000000..3d7fdb6
--- /dev/null
+++ b/vitlarge_feature_cnn_main.py
@@ -0,0 +1,405 @@
+import os
+import pandas as pd
+import nibabel as nib
+import numpy as np
+import torch
+import torch.nn as nn
+import torch.nn.init as init
+import torch.optim as optim
+from tqdm import tqdm
+from nibabel.orientations import io_orientation, axcodes2ornt, ornt_transform, apply_orientation
+from torchvision import transforms
+from transformers import ViTFeatureExtractor, ViTModel
+from sklearn.model_selection import train_test_split
+from sklearn.preprocessing import StandardScaler
+import pickle
+import matplotlib.pyplot as plt
+import sys
+import SimpleITK as sitk
+
+def set_random_seed(seed=69420):
+    np.random.seed(seed)
+    torch.manual_seed(seed)
+    torch.cuda.manual_seed(seed)
+    torch.cuda.manual_seed_all(seed)  # For multi-GPU setups
+    torch.backends.cudnn.deterministic = True
+    torch.backends.cudnn.benchmark = False
+
+set_random_seed(69420)
+
+# Function to resample the volume to 160 slices
+def resample_volume_to_fixed_slices(data, affine, target_slices=160):
+    # Convert Numpy array and affine to SimpleITK image
+    sitk_img = sitk.GetImageFromArray(data)
+    sitk_img.SetSpacing([affine[0, 0], affine[1, 1], affine[2, 2]])
+
+    original_size = sitk_img.GetSize()  # (width, height, depth)
+    original_spacing = sitk_img.GetSpacing()  # (spacing_x, spacing_y, spacing_z)
+
+    # Calculate new spacing to achieve the target number of slices
+    new_spacing = list(original_spacing)
+    new_spacing[2] = (original_spacing[2] * original_size[2]) / target_slices
+
+    # Define new size
+    new_size = [original_size[0], original_size[1], target_slices]
+
+    # Resample the image
+    resampler = sitk.ResampleImageFilter()
+    resampler.SetOutputSpacing(new_spacing)
+    resampler.SetSize(new_size)
+    resampler.SetInterpolator(sitk.sitkLinear)
+    resampled_img = resampler.Execute(sitk_img)
+
+    return sitk.GetArrayFromImage(resampled_img)  # Return the resampled image as a numpy array
+
+# Check if GPU is available
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+# Load the CSV file into a pandas DataFrame
+csv_path = "adni_storage/adni_brainrotnet_metadata.csv"
+df = pd.read_csv(csv_path)
+# df = df.sample(n=1000, random_state=69420)
+print (df)
+# Add a new column 'filepath' with the constructed file paths
+df['filepath'] = df.apply(
+    lambda row: f"adni_storage/ADNI_nii_gz_bias_corrected/I{row['ImageID']}_{row['SubjectID']}.stripped.N4.nii.gz",
+    axis=1
+)
+ 
+
+# Load pre-trained ViT model , vit-large-patch16-224
+feature_extractor = ViTFeatureExtractor.from_pretrained("google/vit-large-patch16-224")
+model = ViTModel.from_pretrained("google/vit-large-patch16-224")
+model.to(device)  # Move the model to the GPU (if available)
+model.eval()
+
+# Update image transform for grayscale images to match ViT input requirements
+transform = transforms.Compose([
+    transforms.ToPILImage(),
+    transforms.Resize((224, 224)),
+    transforms.Lambda(lambda img: img.convert("RGB")),  # Convert to RGB directly
+    transforms.ToTensor(),
+    transforms.Normalize(mean=feature_extractor.image_mean, std=feature_extractor.image_std),
+])
+
+# Directory to save processed images and features
+os.makedirs("adni_storage/ADNI_features_large", exist_ok=True)
+
+# To store features and labels
+features_list = []
+labels_list = []
+
+# Process each row in the DataFrame
+for _, row in tqdm(df.iterrows(), total=len(df), desc="Processing images"):
+    filepath = row['filepath']
+    image_title = f"{row['ImageID']}_{row['SubjectID']}"
+
+    # Check if the feature file already exists
+    feature_file_path = f"adni_storage/ADNI_features_large/{image_title}_features.npy"
+    if os.path.exists(feature_file_path):
+        # If file exists, load the features from the file
+        features = np.load(feature_file_path)
+        features_list.append(features)  # Flatten the features and add to the list
+        labels_list.append(row['Age'])  # Add the corresponding age label
+    else:
+        # print ("hiii")
+        if os.path.exists(filepath):
+            try:
+                # Load the NIfTI image
+                nii_img = nib.load(filepath)
+
+                # Get current orientation and reorient to RAS
+                orig_ornt = io_orientation(nii_img.affine)
+                ras_ornt = axcodes2ornt(("R", "A", "S"))
+                ornt_trans = ornt_transform(orig_ornt, ras_ornt)
+
+                data = nii_img.get_fdata()  # Load image data
+                data = apply_orientation(data, ornt_trans)
+
+                affine = nii_img.affine  # Affine transformation matrix
+
+                # Resample the volume to 160 slices (if required)
+                data = resample_volume_to_fixed_slices(data, affine, target_slices=160)
+
+                # Extract features for all sagittal slices
+                features = []
+                for slice_idx in range(data.shape[0]):
+                    slice_data = data[slice_idx, :, :]
+                    slice_data = (slice_data - np.min(slice_data)) / (np.max(slice_data) - np.min(slice_data))  # Normalize
+
+                    # Transform slice for ViT input
+                    slice_tensor = transform(slice_data).unsqueeze(0).to(device)  # Add batch dimension and move to GPU
+
+                    # Extract features using ViT
+                    with torch.no_grad():
+                        outputs = model(slice_tensor)
+                        slice_features = outputs.last_hidden_state.mean(dim=1).squeeze().cpu().numpy()  # Move output back to CPU
+                        features.append(slice_features)
+
+                # Save extracted features
+                features = np.array(features)
+                np.save(feature_file_path, features)
+                features_list.append(features)
+                labels_list.append(row['Age'])  # Target is 'Age'
+
+            except Exception as e:
+                print(f"Error processing {filepath}: {e}")
+        else:
+            print(f"File not found: {filepath}")
+
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, Dataset
+
+# Custom Dataset
+class ADNIDataset(Dataset):
+    def __init__(self, features_list, sex_list, age_list):
+        self.features = features_list
+        self.sex = sex_list
+        self.age = age_list
+
+    def __len__(self):
+        return len(self.features)
+
+    def __getitem__(self, idx):
+        return (
+            torch.tensor(self.features[idx], dtype=torch.float32),
+            torch.tensor(self.sex[idx], dtype=torch.float32),
+            torch.tensor(self.age[idx], dtype=torch.float32),
+        )
+
+
+# Prepare dataset and dataloaders
+sex_encoded = df['Sex'].apply(lambda x: 0 if x == 'M' else 1).tolist()
+age_list = df['Age'].tolist()
+
+# print (features_list)
+print (features_list[0].shape)
+
+# Create Dataset and DataLoader
+dataset = ADNIDataset(features_list, sex_encoded, age_list)
+train_size = int(0.8 * len(dataset))
+val_size = len(dataset) - train_size
+train_dataset, val_dataset = torch.utils.data.random_split(dataset, [train_size, val_size])
+
+# Store the indices of the validation dataset
+val_indices = val_dataset.indices
+train_indices = train_dataset.indices
+
+train_loader = DataLoader(train_dataset, batch_size=1, shuffle=False)
+val_loader = DataLoader(val_dataset, batch_size=1, shuffle=False)
+
+
+# Tracking outputs for validation samples
+val_outputs = {}
+train_outputs = {}
+
+# Initialize model, loss, and optimizer
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+
+###########################################
+# THIS IS WHERE YOU CHOOSE A MODEL TO TEST 
+###########################################
+
+import importlib
+
+# Assuming sys.argv[1] is the module name
+module_name = sys.argv[1]  # Example: "my_model"
+class_name = "AgePredictionCNN"  # The class you want to import
+
+try:
+    # Dynamically import the module
+    module = importlib.import_module(module_name)
+    
+    # Dynamically get the class
+    AgePredictionCNN = getattr(module, class_name)
+    
+    print(f"Successfully imported {class_name} from {module_name}.")
+
+except ImportError:
+    print(f"Module {module_name} could not be imported.")
+except AttributeError:
+    print(f"{class_name} does not exist in {module_name}.")
+
+##############################
+# MODEL IMPORTED DYNAMICALLY
+##############################
+
+model = AgePredictionCNN(features_list[0].shape).to(device)
+criterion = nn.L1Loss()  # MAE Loss
+optimizer = optim.Adam(model.parameters(), lr=0.007)
+best_loss = np.inf  # Initialize the best loss to infinity
+start_epoch = 0
+
+load_saved = sys.argv[2] # "last, "best"
+if load_saved != "none":
+    # Load the checkpoint
+    with open(f"model_dumps/{sys.argv[1]}_{load_saved}_model_with_metadata.pkl", "rb") as f:
+        checkpoint = pickle.load(f)
+
+    # Restore model and optimizer state
+    model.load_state_dict(checkpoint["model_state"])
+    optimizer.load_state_dict(checkpoint["optimizer_state"])
+
+    # Restore RNG states
+    torch.set_rng_state(checkpoint["t_rng_st"])
+    np.random.set_state(checkpoint["n_rng_st"])
+    if torch.cuda.is_available() and checkpoint["cuda_rng_st"] is not None:
+        torch.cuda.set_rng_state_all(checkpoint["cuda_rng_st"])
+
+    # Retrieve metadata
+    start_epoch = checkpoint["epoch"] + 1
+    loaded_loss = checkpoint["loss"]
+
+    print(f"Loaded model from epoch {start_epoch} with best validation loss: {loaded_loss:.4f}")
+
+predicted_ages = None
+# Training loop
+epochs = 200
+
+# Initialize lists to track loss
+filename = sys.argv[1] 
+csv_file = f"model_dumps/{filename}.csv"
+
+# Load existing epoch data if the file exists
+if os.path.exists(csv_file):
+    epoch_data = pd.read_csv(csv_file).to_dict(orient="records")
+    print(f"Loaded existing epoch data from {csv_file}.")
+else:
+    epoch_data = []
+    print("No existing epoch data found. Starting fresh.")
+
+
+# Plot loss vs. epoch and save the figure
+def update_loss_plot(epoch_data, filename):
+    df = pd.DataFrame(epoch_data)
+    df.to_csv(f"model_dumps/{filename}.csv", index=False)  # Save the data to CSV
+    
+    plt.figure(figsize=(8, 6))
+    plt.plot(df['epoch'], df['train_loss'], label="Train Loss", marker="o")
+    plt.plot(df['epoch'], df['val_loss'], label="Validation Loss", marker="o")
+    plt.xlabel("Epoch")
+    plt.ylabel("Loss")
+    plt.title("Loss vs. Epoch")
+    plt.legend()
+    plt.grid(True)
+    plt.savefig(f"model_dumps/{filename}.png")
+    plt.close()
+
+# Training loop
+for epoch in range(start_epoch, epochs):
+    model.train()
+    train_loss = 0.0
+    predicted_ages = []
+
+    for idx, (features, sex, age) in enumerate(train_loader):
+        features = features.unsqueeze(1).to(device)  # Add channel dimension
+        sex = sex.to(device)
+        age = age.to(device)
+        optimizer.zero_grad()
+        outputs = model(features, sex)
+        train_outputs[train_indices[idx]] = outputs.item()
+
+        loss = criterion(outputs.squeeze(), age)
+        loss.backward()
+        optimizer.step()
+
+        train_loss += loss.item()
+
+    train_loss /= len(train_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Train Loss: {train_loss:.4f}")
+
+    # Validation loop
+    model.eval()
+    val_loss = 0.0
+    with torch.no_grad():
+        for idx, (features, sex, age) in enumerate(val_loader):
+            features = features.unsqueeze(1).to(device)
+            sex = sex.to(device)
+            age = age.to(device)
+
+            outputs = model(features, sex)
+            loss = criterion(outputs.squeeze(), age)
+            val_loss += loss.item()
+
+            # Save the predicted age for the current validation sample
+            val_outputs[val_indices[idx]] = outputs.item()
+
+    val_loss /= len(val_loader)
+    print(f"Epoch {epoch+1}/{epochs}, Validation Loss: {val_loss:.4f}")
+
+    # Save the last model with metadata
+    print(f"Saving last model...")
+    checkpoint = {
+        "model_state": model.state_dict(),
+        "optimizer_state": optimizer.state_dict(),
+        "epoch": epoch,
+        "loss": val_loss,
+        "t_rng_st": torch.get_rng_state(),
+        "n_rng_st": np.random.get_state(),
+        "cuda_rng_st": torch.cuda.get_rng_state_all() if torch.cuda.is_available() else None
+    }
+    with open(f"model_dumps/{sys.argv[1]}_last_model_with_metadata.pkl", "wb") as f:
+        pickle.dump(checkpoint, f)
+    print(f"Last model saved...")
+
+    # Check if validation loss improved
+    if val_loss < best_loss:
+        best_loss = val_loss
+        print(f"Validation loss improved to {best_loss:.4f}. Saving model...")
+        with open(f"model_dumps/{sys.argv[1]}_best_model_with_metadata.pkl", "wb") as f:
+            pickle.dump(checkpoint, f)
+
+    # Save predictions and create DataFrames (same as before)
+    # ...
+
+    # Update epoch data and save the loss plot
+    epoch_data.append({
+        "epoch": epoch + 1,
+        "train_loss": train_loss,
+        "val_loss": val_loss
+    })
+    update_loss_plot(epoch_data, sys.argv[1])
+
+    max_index = max(train_outputs.keys())
+    # Create a DataFrame with NaN for all indices initially
+    df_trn = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
+    # Assign the values to their respective indices
+    for index, value in train_outputs.items():
+        df_trn.loc[index, "Predicted_Age"] = value
+    # print (df_trn)
+
+    df2 = df.copy()
+    df2['Predicted_Age'] = df_trn['Predicted_Age']
+    train_df = df2.loc[train_outputs.keys()]
+    print (train_df)
+    train_df.to_csv(f"model_dumps/{sys.argv[1]}_predicted_ages_train.csv")
+
+    max_index = max(val_outputs.keys())
+    # Create a DataFrame with NaN for all indices initially
+    df_pred = pd.DataFrame(index=range(max_index + 1), columns=["Predicted_Age"])
+    # Assign the values to their respective indices
+    for index, value in val_outputs.items():
+        df_pred.loc[index, "Predicted_Age"] = value
+    # print (df_pred)
+
+    df1 = df.copy()
+    df1['Predicted_Age'] = df_pred['Predicted_Age']
+    test_df = df1.loc[val_outputs.keys()]
+    # print (test_df)
+    test_df.to_csv(f"model_dumps/{sys.argv[1]}_predicted_ages_val.csv")
+
+
+    # Check that the predictions have been added to the DataFrame
+    # Plot Age vs. Predicted Age
+    # plt.figure(figsize=(8, 6))
+    # plt.scatter(train_df['Age'], train_df['Predicted_Age'], color='blue', label='Predicted vs Actual')
+    # # plt.plot(test_df['Age'], test_df['Age'], color='red', linestyle='--', label='Perfect Prediction')  # Optional: Line of perfect prediction
+    # plt.xlabel('Age')
+    # plt.ylabel('Predicted Age')
+    # plt.title('Age vs Predicted Age')
+    # plt.legend()
+    # plt.grid(True)
+    # plt.show()
\ No newline at end of file