3DIRCADb Vessel+

This dataset provides extended vascular annotations for the 3DIRCADb liver CT dataset, designed to support research in hepatic vessel segmentation using deep learning. Each of the 20 patients in the original 3DIRCADb dataset is associated with a single binary mask in NIfTI format (`mask.nii.gz`) that combines the major hepatic vascular structures, including hepatic arteries, portal vein, hepatic veins, and the inferior vena cava. The annotations are fully anonymized and were manually constructed and post-processed to ensure consistency across patients. The folder structure mirrors the original 3DIRCADb dataset (e.g., patient01, patient02...) to facilitate easy integration. The original CT volumes are not included and must be downloaded from IRCAD’s official website: https://www.ircad.fr/research/3dircadb/ The annotations are distributed under a CC BY 4.0 license.