TotalSpineSeg

TotalSpineSeg is a tool for automatic instance segmentation and labeling of all vertebrae, intervertebral discs (IVDs), spinal cord, and spinal canal in MRI images. It follows the TotalSegmentator classes with additional classes for IVDs, spinal cord, and spinal canal (see list of classes). The model is based on nnUNet as the backbone for training and inference.

If you use this model, please cite our work:

Warszawer Y, Molinier N, Valošek J, Shirbint E, Benveniste PL, Achiron A, Eshaghi A and Cohen-Adad J. Fully Automatic Vertebrae and Spinal Cord Segmentation Using a Hybrid Approach Combining nnU-Net and Iterative Algorithm. Proceedings of the 32th Annual Meeting of ISMRM. 2024

Please also cite nnUNet since our work is heavily based on it:

Isensee, F., Jaeger, P. F., Kohl, S. A., Petersen, J., & Maier-Hein, K. H. (2021). nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation. Nature methods, 18(2), 203-211.

Table of Contents

• Model Description
• Datasets
• Dependencies
• Installation
• Training
• Inference
• Output Examples
• List of Classes

Model Description

TotalSpineSeg uses a hybrid approach that integrates nnU-Net with an iterative algorithm for instance segmentation and labeling of vertebrae, intervertebral discs (IVDs), spinal cord, and spinal canal. The process involves two main steps:

Step 1: An nnUnet model (Dataset101) was trained to identify 8 classes in total (Figure 1A). This includes 4 main classes: spinal cord, spinal canal, IVDs, and vertebrae. Additionally, it identifies 4 specific IVDs: C2-C3, C7-T1, T12-L1, and L5-S, which represent key anatomical landmarks along the spine. The output segmentation was then processed using an iterative algorithm. This algorithm extracts odd IVDs segmentation based on the C2-C3, C7-T1, T12-L1, and L5-S IVD labels produced by the model (Figure 1B).

Step 2: A second nnUNet model (Dataset102) was trained to identify 14 classes in total (Figure 1C). This includes 6 main classes: spinal cord, spinal canal, odd IVDs, even IVDs, odd vertebrae, and even vertebrae. Additionally, it identifies 4 specific IVDs: C2-C3, C7-T1, T12-L1, and L5-S, and 4 specific vertebrae: C2, T1, T12, and Sacrum. This model uses two input channels: the MRI image and the odd IVDs extracted from the first step. The output segmentation was then processed using an algorithm that assigns an individual label value to each vertebra and IVD in the final segmentation mask (Figure 1D).

For comparison, we also trained a single model (Dataset103) that outputs individual label values for each vertebra and IVD in a single step.

Figure 1: Illustration of the hybrid method for automatic segmentation of the spine and spinal cord structures. T1w image (A) is used to train model 1, which outputs 8 classes (B). These output labels are processed to extract odd IVDs (C). The T1w and odd IVDs are used as two input channels to train model 2, which outputs 14 classes (D). These output labels are processed to extract individual IVDs and vertebrae (E).

Datasets

The totalspineseg model was trained on these 3 main datasets:

• Private whole-spine dataset (Internal access: git@data.neuro.polymtl.ca:datasets/whole-spine.git).
• SPIDER project dataset (Internal access: git@data.neuro.polymtl.ca:datasets/spider-challenge-2023.git)
• Spine Generic Project, including single and multi subject datasets (Public access: git@github.com:spine-generic/data-single-subject.git and git@github.com:spine-generic/data-multi-subject.git).

Additional public datasets were used during this project to generate sacrum segmentations:

• GoldAtlas (Internal access: git@data.neuro.polymtl.ca:datasets/goldatlas.git)
• SynthRAD2023 (Internal access: git@data.neuro.polymtl.ca:datasets/synthrad-challenge-2023.git)
• MRSpineSeg (Internal access: git@data.neuro.polymtl.ca:datasets/mrspineseg-challenge-2021.git)

When not available, sacrum segmentations were generated using the totalsegmentator model. For more information, please see this issue.

Dependencies

• bash terminal
• Python >= 3.9, with pip >= 23 and setuptools >= 67

Installation

1. Open a bash terminal in the directory where you want to work.

2. Create the installation directory:

   mkdir TotalSpineSeg
   cd TotalSpineSeg

3. Create and activate a virtual environment (highly recommended):

   python3 -m venv venv
   source venv/bin/activate

4. Clone and install this repository:

   git clone https://github.com/neuropoly/totalspineseg.git
   python3 -m pip install -e totalspineseg

5. For CUDA GPU support, install PyTorch following the instructions on their website. Be sure to add the --upgrade flag to your installation command to replace any existing PyTorch installation. Example:

   python3 -m pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118 --upgrade

6. Set the path to TotalSpineSeg and data folders in the virtual environment:

   mkdir data
   export TOTALSPINESEG="$(realpath totalspineseg)"
   export TOTALSPINESEG_DATA="$(realpath data)"
   echo "export TOTALSPINESEG=\"$TOTALSPINESEG\"" >> venv/bin/activate
   echo "export TOTALSPINESEG_DATA=\"$TOTALSPINESEG_DATA\"" >> venv/bin/activate

Training

To train the TotalSpineSeg model, you will need the following hardware specifications:

• Approximately 3.5TB of available disk space (for training with data augmentation)
• RAM capacity of at least 32GB
• CUDA GPU with at least 8GB of VRAM

Please ensure that your system meets these requirements before proceeding with the training process.

1. Make sure that the bash terminal is opened with the virtual environment (if used) activated (using source <path to installation directory>/venv/bin/activate).

2. Ensure training dependencies are installed:

   apt-get install git git-annex jq -y

3. Download the required datasets into $TOTALSPINESEG_DATA/bids (make sure you have access to the specified repositories):

   bash "$TOTALSPINESEG"/scripts/download_datasets.sh

4. Temporary step (until all labels are pushed into the repositories) - Download labels into $TOTALSPINESEG_DATA/bids:

   curl -L -O https://github.com/neuropoly/totalspineseg/releases/download/labels/labels_iso_bids_0524.zip
   unzip -qo labels_iso_bids_0524.zip -d "$TOTALSPINESEG_DATA"
   rm labels_iso_bids_0524.zip

5. Prepare datasets in nnUNetv2 structure into $TOTALSPINESEG_DATA/nnUnet:

   bash "$TOTALSPINESEG"/scripts/prepare_datasets.sh [DATASET_ID] [-noaug]

   The script optionally accepts DATASET_ID as the first positional argument to specify the dataset to prepare. It can be either 101, 102, 103, or all. If all is specified, it will prepare all datasets (101, 102, 103). By default, it will prepare datasets 101 and 102.

   Additionally, you can use the -noaug parameter to prepare the datasets without data augmentations.

6. Train the model:

   bash "$TOTALSPINESEG"/scripts/train.sh [DATASET_ID [FOLD]]

   The script optionally accepts DATASET_ID as the first positional argument to specify the dataset to train. It can be either 101, 102, 103, or all. If all is specified, it will train all datasets (101, 102, 103). By default, it will train datasets 101 and 102.

   Additionally, you can specify FOLD as the second positional argument to specify the fold. It can be either 0, 1, 2, 3, 4, 5 or all. By default, it will train with fold 0.

Inference

1. Make sure that the bash terminal is opened with the virtual environment (if used) activated (using source <path to installation directory>/venv/bin/activate).

2. Run the model on a folder containing the images in .nii.gz format:

   totalspineseg INPUT_FOLDER OUTPUT_FOLDER [-step1]

   This will process all .nii.gz files in the INPUT_FOLDER and save the results in the OUTPUT_FOLDER. If you haven't trained the model, the script will automatically download the pre-trained models from the GitHub release.

   Additionally, you can use the -step1 parameter to run only the step 1 model, which outputs a single label for all vertebrae, including the sacrum.

Output Examples

TotalSpineSeg demonstrates robust performance across a wide range of imaging parameters. Here are some examples of the model output:

The examples shown above include segmentation results on various contrasts (T1w, T2w, STIR, MTS, T2star, and even CT images), acquisition orientations (sagittal, axial), and resolutions.

For a more detailed view of the output examples, you can check the PDF version:

Preview PDF

The PDF includes step 1 and step 2 results together with the iterative labeling algorithm for each step.

List of Classes

Label	Name
18	vertebrae_L5
19	vertebrae_L4
20	vertebrae_L3
21	vertebrae_L2
22	vertebrae_L1
23	vertebrae_T12
24	vertebrae_T11
25	vertebrae_T10
26	vertebrae_T9
27	vertebrae_T8
28	vertebrae_T7
29	vertebrae_T6
30	vertebrae_T5
31	vertebrae_T4
32	vertebrae_T3
33	vertebrae_T2
34	vertebrae_T1
35	vertebrae_C7
36	vertebrae_C6
37	vertebrae_C5
38	vertebrae_C4
39	vertebrae_C3
40	vertebrae_C2
41	vertebrae_C1
92	sacrum
200	spinal_cord
201	spinal_canal
202	disc_L5_S
203	disc_L4_L5
204	disc_L3_L4
205	disc_L2_L3
206	disc_L1_L2
207	disc_T12_L1
208	disc_T11_T12
209	disc_T10_T11
210	disc_T9_T10
211	disc_T8_T9
212	disc_T7_T8
213	disc_T6_T7
214	disc_T5_T6
215	disc_T4_T5
216	disc_T3_T4
217	disc_T2_T3
218	disc_T1_T2
219	disc_C7_T1
220	disc_C6_C7
221	disc_C5_C6
222	disc_C4_C5
223	disc_C3_C4
224	disc_C2_C3