Brain Tumor Detection Model Diagnostic 🧠

Project Description

This project uses a machine learning model to identify brain tumors based on MRI images. It supports classification into four categories:

1. Pituitary Tumor
2. No Tumor
3. Meningioma
4. Glioma

The model is deployed on Streamlit and provides an intuitive interface for users to upload MRI images, view predictions, and analyze probabilities for each class.

Features

• Upload and test MRI images.
• View predictions with confidence probabilities.
• Intuitive UI hosted on Streamlit.
• Real-time processing and results display.

Hosted Link

You can access the project here.

---

How It Works

1. Upload an MRI image (JPG, PNG, JPEG).
2. The image is preprocessed and passed through a pre-trained model.
3. The model predicts the tumor type or indicates if no tumor is present.
4. Results and probabilities are displayed instantly.

---

How to Run Locally

1. Clone the repository:

   git clone https://github.com/Sandeep0900/brain-tumor-detection.git
   cd brain-tumor-detection

2. Install the required packages:

   pip install -r requirements.txt

3. Place the keras_model.h5 file in the models directory.
4. Run the Streamlit app:

   streamlit run app.py

5. Open the app in your browser at http://localhost:8501.

---

Future Enhancements

• Improve model accuracy with additional data.
• Add more tumor classifications.
• Deploy on additional platforms for broader accessibility.

Feel free to explore, contribute, and give feedback. Let's make diagnostics smarter together! 🚀