This is the *subject based structure* Datastructure of VerSe2020. Please download and read the readme.txt and licence.txt of VerSe2020. This is the extension of VerSe2019 which is available in the same format here: https://osf.io/923ap/ and can be used together.
In the *subject based structure*, subdirectories for each subject/patient as well as a split in rawdata and derivatives were introduced. Directory names are subject identifiers. File names are constructed of entities, a suffix and a file extension following the conventions of the Brain Imaging Data Structure (BIDS; https://bids.neuroimaging.io/). For patients with multiple image series, we included a 'split-<value>' entity, where <value> represents the original image series identifier.
Example:
training/rawdata/sub-verse000/
(1) sub-verse000_dir-orient_ct.nii.gz - CT image series
training/derivatives/sub-verse000/
(2) sub-verse000_dir-orient_seg-vert_msk.nii.gz - segmentation mask of the vertebrae
(3) sub-verse000_dir-orient_seg-subreg_ctd.json - centroid coordinates in image space
(4) verse000_dir-orient_seg-vert_snp.png - Preview reformations of the annotated CT data.
Centroid coordinates of the *subject based structure* (.json file) are given in voxels in the image space.
'label' corresponds to the vertebral label:
1-7: cervical spine: C1-C7
8-19: thoracic spine: T1-T12
20-25: lumbar spine: L1-L6
26,27: sacrum, cocygis - not labeled in this dataset
28: additional 13th thoracic vertebra, T13
The dataset has been prepared for the MICCAI 2020 challenge "VerSe": https://verse2020.grand-challenge.org/
The data is published under the licence CC BY-SA 4.0 (see licence.txt, https://osf.io/amh4f/). When using the data you must cite the three papers mentioned in readme.txt.
