๐Ÿฉป BTCV โ€” Beyond the Cranial Vault

English็ฎ€ไฝ“ไธญๆ–‡

Abdominal multi-organ segmentation ยท CT

Official dataset source Related challenge

๐Ÿ“š Publications and records

MICCAI workshop 2015 Dataset DOI 10.7303/syn3193805

Quick facts
Input Abdominal CT
Targets 13 organs
Institutional origin The abdominal CT scans were acquired at Vanderbilt University Medical Center. The dataset was released through the MICCAI 2015 Multi-Atlas Labeling Beyond the Cranial Vault workshop challenge. Challenge record.
Challenge source MICCAI 2015 Multi-Atlas Labeling Beyond the Cranial Vault

๐Ÿงพ Dataset overview

BTCV is associated with the MICCAI 2015 Multi-Atlas Labeling Beyond the Cranial Vault workshop and challenge. The abdominal challenge data contain 30 labeled CT cases for development and a separate 20-case test cohort. The labels cover 13 organs: spleen, right and left kidneys, gallbladder, esophagus, liver, stomach, aorta, inferior vena cava, portal and splenic veins, and right and left adrenal glands.

๐Ÿ—‚๏ธ Official split and label availability

Cohort Cases Labels
Training/development 30 Reference labels provided
Test 20 Test labels are withheld for challenge scoring

The official source defines training and test cohorts; it does not establish the commonly used 24/6 or 18/12 training/validation splits as organizer-defined splits. Those are secondary experimental protocols and are intentionally not listed as official here.

๐Ÿ”’ Access and terms

MICCAI 2015 Multi-Atlas Labeling Beyond the Cranial Vault was a workshop challenge for multi-organ abdominal CT labeling; the BTCV abdomen cohort is the 13-structure abdominal task.

๐Ÿ“š Publication and citation

Landman et al. โ€œMICCAI Multi-Atlas Labeling Beyond the Cranial Vault โ€” Workshop and Challenge.โ€ Proceedings of the MICCAI Multi-Atlas Labeling Beyond the Cranial Vault Workshop and Challenge, 2015, vol. 5, p. 12.

Use the challenge dataset record for access details and the workshop citation when referencing the original benchmark.


๐Ÿ•˜ Sources last checked: 27 October 2026 ยท Back to the dataset catalog