All datasets
AvailableCT

VerSe

VerSe is a multi-centre spine CT benchmark created for vertebra identification, labelling and voxel-level segmentation. It combines the datasets released for the VerSe 2019 and VerSe 2020 challenges.

Licence and access

The restructured VerSe 2019 and VerSe 2020 data are openly distributed under the Creative Commons Attribution-ShareAlike 4.0 International license. The legacy Grand Challenge page refers to CC BY-SA 2.0, while the current official repository specifies CC BY-SA 4.0.

open accesscommercial use allowedredistribution allowed
Visit official sourcelast verified 2026-09-03
Coverage
spinemixed labelsin vivo
subjects
355
images
374
formats
NIfTI, JSON
origin
Germany · 2020
Citation
Sekuboyina A, Husseini ME, Bayat A, et al. VerSe: A vertebrae labelling and segmentation benchmark for multi-detector CT images. Medical Image Analysis. 2021;73:102166. DOI: 10.1016/j.media.2021.102166.
Notes

The combined VerSe 2019 and 2020 datasets contain 374 CT scans from 355 patients and 4,505 individually annotated vertebrae. Segmentation masks are provided in NIfTI format, while vertebral centroid annotations are provided in JSON files.

This record is a pointer. Licence and access terms are recorded as they stood on the verification date above; what binds you is the dataset's own licence text. Confirm at the source before use.

Orthopedic AI Research Updates

Monthly research digest, product updates, and clinical AI insights.

Unsubscribe anytime.

VerSe, dataset record