AbdomenAtlas/AbdomenAtlasNet

This repository contains two segmentation models trained on AbdomenAtlas 3.0. Both are nnU-Nets.

AbdomenAtlasNet-Tumors segments:

• tumors in the liver, kidney, and pancreas

AbdomenAtlasNet-Organs segments:

• blood vessels, including the aorta, IVC, SMA, CHA, CA, SA, portal vein, and splenic vein
• liver sub-segments & pancreas sub-segments
• 18 organs, including the gallbladder, left kidney, right kidney, spleen, stomach, left adrenal gland, right adrenal gland, bladder, colon, duodenum, esophagus, left femur, right femur, intestine, left lung, right lung, prostate, and rectum

Instructions

Both models are nnU-Net V2 checkpoints. For use intructions, please click here.

Note: these models were trained on the whole AbdomenAtlas 3.0 dataset. Therefore, do not test them on AbdomenAtlas 3.0, since you will be testing on the training set.

Dataset

AbdomenAtlas 3.0 is the first public dataset with high quality abdominal CTs and paired radiology reports. The database includes more than 9,000 CT scans with radiology reports and per-voxel annotations of liver, kidney and pancreatic tumors.

You can download AbdomenAtlas 3.0 here:

• AbdomenAtlas 3.0

Paper

RadGPT: Constructing 3D Image-Text Tumor Datasets
Pedro R. A. S. Bassi, Mehmet Yavuz, Kang Wang, Sezgin Er, Ibrahim E. Hamamci, Wenxuan Li, Xiaoxi Chen, Sergio Decherchi, Andrea Cavalli, Yang Yang, Alan Yuille, Zongwei Zhou*
Johns Hopkins University
ICCV, 2025

Citation

@article{bassi2025radgpt,
  title={RadGPT: Constructing 3D Image-Text Tumor Datasets},
  author={Bassi, Pedro RAS and Yavuz, Mehmet Can and Wang, Kang and Chen, Xiaoxi and Li, Wenxuan and Decherchi, Sergio and Cavalli, Andrea and Yang, Yang and Yuille, Alan and Zhou, Zongwei},
  journal={arXiv preprint arXiv:2501.04678},
  year={2025},
  url={https://github.com/MrGiovanni/RadGPT}
}

Acknowledgement

This work was supported by the Lustgarten Foundation for Pancreatic Cancer Research, the Patrick J. McGovern Foundation Award, and the National Institutes of Health (NIH) under Award Number R01EB037669. We would like to thank the Johns Hopkins Research IT team in IT@JH for their support and infrastructure resources where some of these analyses were conducted; especially DISCOVERY HPC. Paper content is covered by patents pending.