Model repository for MS lesion segmentation on MP2RAGE data (UNIT1 contrast)
3D model trained with nnUNetv2 framework.
basel-mp2rage
nih-ms-mp2rage
marseille-3t-mp2rage
sct_deepseg -install-task seg_sc_contrast_agnostic
sct_deepseg -install-task seg_ms_lesion_mp2rage
Warning: When running the MS lesion segmentation model, the image first need to be cropped around the spinal cord mask with a dilation of 30 mm in the axial plane and 5 mm in the Z-axis.
sct_deepseg -i IMAGE_UNIT1 -task seg_sc_contrast_agnostic -o IMAGE_UNIT1_sc
sct_crop_image -i IMAGE_UNIT1 -m IMAGE_UNIT1_sc -dilate 30x30x5 -o IMAGE_UNIT1_cropped
sct_deepseg -i IMAGE_UNIT1_cropped -task seg_ms_lesion_mp2rage
Download and unzip the nnUNetTrainer_seg_ms_lesion_mp2ragennUNetPlans3d_fullres.zip file. (~120 Mb)
Unzip the .zip
file and place it inside a folder named Dataset403_seg_ms_lesion_mp2rage_1mm_322subj
. The final directory structure should look like this:
Dataset403_seg_ms_lesion_mp2rage_1mm_322subj
└── nnUNetTrainer_seg_ms_lesion_mp2rage__nnUNetPlans__3d_fullres
├── dataset_fingerprint.json
├── dataset.json
├── dataset_split.md
├── datasplits
│ ├── datasplit_basel-mp2rage.yaml
│ ├── datasplit_marseille-3t-mp2rage.yaml
│ └── datasplit_nih-ms-mp2rage.yaml
├── fold_3
│ ├── checkpoint_final.pth
│ ├── debug.json
│ ├── progress.png
│ └── training_log_2024_3_14_20_35_08.txt
└── plans.json
nnUNet Install: Follow the instructions on first row of:
Implementation
Model path
model_seg_ms_mp2rage
!! 🚀🚀🚀