Skip to content

Latest commit

 

History

History
93 lines (68 loc) · 3.97 KB

README.md

File metadata and controls

93 lines (68 loc) · 3.97 KB
Raw

model_seg_ms_mp2rage

release zenodo

Model repository for MS lesion segmentation on MP2RAGE data (UNIT1 contrast)

Model overview

uniseg_update

3D model trained with nnUNetv2 framework.

Dependencies

SCT

Datasets for training

  • basel-mp2rage
  • nih-ms-mp2rage
  • marseille-3t-mp2rage

Installation

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.

Launch MS lesion segmentation:

  1. Spinal cord segmentation
sct_deepseg -i IMAGE_UNIT1 -task seg_sc_contrast_agnostic -o IMAGE_UNIT1_sc
  1. Cropping with dilation (for axial orientation images with 1mm isotropic resolution)
sct_crop_image -i IMAGE_UNIT1 -m IMAGE_UNIT1_sc -dilate 30x30x5 -o IMAGE_UNIT1_cropped
  1. MS lesion segmentation
sct_deepseg -i IMAGE_UNIT1_cropped -task seg_ms_lesion_mp2rage

Model implementation on Slicer

  1. Install 3D Slicer version 5.7.0, then install the module SlicerNNUnet from the extensions explorer.

  1. Download and unzip the nnUNetTrainer_seg_ms_lesion_mp2rage__nnUNetPlans__3d_fullres.zip file. (~120 Mb)

  2. 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

  1. nnUNet Install: Follow the instructions on first row of: Slicer-UNIseg

  2. Implementation

  • Navigate and select the Slicer friendly folder on Model path
  • Folds: Choose the fold 3
  • Apply model_seg_ms_mp2rage !! 🚀🚀🚀

Acknowledgments

  • Charidimos Tsagkas (Translational Neuroradiology Section, National Institutes of Health, Bethesda, USA)
  • Daniel Reich (Translational Neuroradiology Section, National Institutes of Health, Bethesda, USA)
  • Cristina Granziera (Neurologic Clinic and Policlinic, University Hospital Basel and University of Basel, Basel, Switzerland)
  • Virginie Callot (Center for Magnetic Resonance in Biology and Medicine, CRMBM-CEMEREM, UMR 7339, CNRS, Aix-Marseille University, Marseille, France)

How to cite

Laines Medina N, Mchinda S, Testud B, Demortière S, Chen M, Granziera G, Reich D, Tsagkas C, Cohen-Adad J and Callot V. Automatic Multiple Sclerosis Lesion Segmentation in the Spinal Cord on 3T and 7T MP2RAGE images, in: Proceedings of the 40th Annual Scientific Meeting of ESMRMB Barcelona, Spain 2024 p171.