Multi-Organ Abdominal CT Reference Standard Segmentations for 90 Images

Eli Gibson and University College London released this dataset of reference segmentations for 90 abdominal CT images. The data delineates multiple organs, including the spleen, kidneys, gallbladder, esophagus, liver, stomach, pancreas, and duodenum. It combines and extends segmentations from The Cancer Image Archive Pancreas-CT dataset and the Beyond the Cranial Vault Abdomen dataset.

Use Cases

• Training semantic segmentation models for abdominal organs based on the described reference masks.
• Benchmarking segmentation algorithm performance against a verified medical standard.
• Developing models for organ localization and volumetric analysis in abdominal CT scans.
• Studying anatomical variation across patient cohorts described in the source datasets.

Strengths

• Includes 90 abdominal CT images with reference segmentations.
• Segmentations were manually labeled and verified by radiologists, as described.
• Combines data from two established medical imaging sources (TCIA and BTCV).

Limitations

• Row count and dataset size are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

University College London, combining data from The Cancer Image Archive (NIH) and Vanderbilt University Medical Center.

Collection Method

Manual slice-by-slice and volumetric labeling by students and fellows, with verification and editing by experienced radiologists.

Time Range

null

Freshness

null

Geography

null