TotalSpineSeg
Two-stage nnU-Net tool that segments and individually labels every vertebra and intervertebral disc plus the spinal cord and canal, using an odd/even vertebra scheme and landmark discs (C2-C3, C7-T1, T12-L1, L5-S) to avoid fused or split vertebrae; works on localizers and short-FOV scans; peer-reviewed in European Radiology (Nov 7, 2026)
- What it is
- Segmentation (Segmentation & frameworks)
- Modality
- Radiology: spine MRI (any contrast), CT
- Maintainer
- NeuroPoly (Polytechnique Montreal) / Warszawer, Molinier et al.
- License
- LGPL-3.0
- Commercial use
- Copyleft
- Access
- Free, public source repository (see links below)
- Last update
- 2026-08-10 (last repository push, UTC; Maintained)
- Last checked
- 2026-11-16
- Limitations
- Listed for research and development reference. Hounsfield Lab has not tested its clinical performance; check the project’s own documentation for intended use and regulatory status.
Links
License review: Repo LICENSE is GNU LGPL v3. Trained nnU-Net weights (r20260730) are GitHub release assets in the same repository and are downloaded on first run; no separate weight license is stated. Trained on 1,404 MRIs; the paper reports 96.22% vertebral identification on localizer MRI and 92.87% on CT. Paper: European Radiology, doi:10.1007/s00330-026-13008-2.
Commercial use: read the project’s license (and any model-weight terms) yourself before relying on it. This isn’t legal advice.
Related tools in Segmentation & frameworks
- MONAI: PyTorch-based AI toolkit for healthcare imaging
- MONAI Label: AI-assisted image labeling and active learning; integrates with 3D Slicer and OHIF
- nnU-Net: Self-configuring segmentation framework
- TotalSegmentator: Segments more than 100 anatomical structures in CT and MR
- CXAS: Chest X-ray anatomy segmentation (Mask2Former models: 187 frontal and 177 lateral classes) plus 132 named landmarks and landmark-based measurements such as cardiothoracic ratio, costophrenic angle, Cobb angle and the Haller index, in mm for DICOM input; cxas_idc writes one multi-segment DICOM SEG and a TID 1500 SR per radiograph; trained on PAXRay++ CT-projected radiographs
- NV-Segment-CT: CT segmentation built from MONAI VISTA3D: 117 structures automatically, or 132 promptable by point or class. Research use only, not for clinical care. Needs a CUDA GPU; weights are about 872 MB
Related public imaging datasets
- Spine-Mets-CT-SEG: CT, MSK · CC BY 4.0
- TotalSegmentator CT dataset: CT, Body, Chest, MSK · CC BY 4.0
- MRNet: MR, MSK · Stanford AIMI research use agreement (non-commercial; commercial license sold separately)
- NYU fastMRI: MR, MSK, Neuro, Body, Breast · NYU fastMRI Dataset Sharing Agreement (internal research or education only)
- RSNA 2022 Cervical Spine Fracture Detection: CT, MSK · Kaggle competition rules (non-commercial)
- RSNA 2024 Lumbar Spine Degenerative Classification: MR, MSK · Kaggle competition rules (non-commercial)
Browse all DICOM datasets with license and access details.