Once the setup is complete, users can simply click on medseg_exe.py to run MedSeg on their laptop. The browser will automatically open with the appropriate URL, taking users directly to the MedSeg login page.
Users should enter their username and password to access MedSeg.
- Click on the green "Add Project" button.
- Enter your desired project name.
- Press "OK".
- Click on the blue "Upload" button.
- Select the images you wish to upload.
Important:
-
If you're using an image as a primary image or as a mask, their filenames must differ.
- For a primary image, the naming format should be:
img_MYIMAGE - For the mask of this image, it should be:
msk_MYIMAGE.
"MYIMAGE" can be replaced with any desired name, but ensure the identifier remains consistent between an image and its corresponding mask.
- For a primary image, the naming format should be:
-
Files must be in the
.nii.gzextension. (nifti format ONLY) -
You can put many images at the same time
Below are the steps and visual examples for uploading images and their corresponding masks:
-
Uploading Your First Image:
Ensure that the image has the correct name and extension.
-
Verifying the Image Upload:
On your dashboard, you should be able to see that the image has been added successfully.
-
Adding the Corresponding Mask:
The mask should have the same name as the image but with amskprefix.
Users can upload both the image and its mask simultaneously. When uploading, users can categorize the mask as 'train', 'validate', or 'upsampling'. If not specified, the mask will default to '1', but it can be renamed later.
-
Mask Upload Verification:
A small downward arrow (v) in the segmentation column indicates that the mask has been uploaded correctly for the corresponding image.
