Automatic Segmentation of Spinal Nerve Rootlets

This repository contains the code for deep learning-based segmentation of the spinal nerve rootlets. The code is based on the nnUNet framework.

Citation Info

If you find this work and/or code useful for your research, please refer to the following preprint:

@misc{valosek2024automatic,
      title={Automatic Segmentation of the Spinal Cord Nerve Rootlets}, 
      author={Jan Valosek and Theo Mathieu and Raphaelle Schlienger and Olivia S. Kowalczyk and Julien Cohen-Adad},
      year={2024},
      eprint={2402.00724},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}

Model Overview

The model was trained on T2-weighted images and provides semantic (i.e., level-specific) segmentation of the dorsal spinal nerve rootlets.

How to use the model

Install dependencies

• Spinal Cord Toolbox (SCT) v6.2 or higher -- follow the installation instructions here
• conda
• Python

Once the dependencies are installed, download the latest rootlets model:

sct_deepseg -install-task seg_spinal_rootlets_t2w

Getting the rootlet segmentation

To segment a single image, run the following command:

sct_deepseg -i <INPUT> -o <OUTPUT> -task seg_spinal_rootlets_t2w

For example:

sct_deepseg -i sub-001_T2w.nii.gz -o sub-001_T2w_label-rootlets_dseg.nii.gz -task seg_spinal_rootlets_t2w