Datasets:

mrmrx
/

CADS-dataset

@article{ji2022amos,
  title={Amos: A large-scale abdominal multi-organ benchmark for versatile medical image segmentation},
  author={Ji, Yuanfeng and Bai, Haotian and Ge, Chongjian and Yang, Jie and Zhu, Ye and Zhang, Ruimao and Li, Zhen and Zhanng, Lingyan and Ma, Wanling and Wan, Xiang and others},
  journal={Advances in neural information processing systems},
  volume={35},
  pages={36722--36732},
  year={2022}
}

Dataset description

AMOS is a large-scale abdominal multi-organ segmentation benchmark designed to advance clinical applications such as disease diagnosis and treatment planning. It contains 500 CT and 100 MRI scans from multi-center, multi-vendor, multi-modality, and multi-phase acquisitions, covering patients with a variety of abdominal diseases. Each case includes voxel-level annotations for 15 abdominal organs, enabling the development and fair comparison of versatile segmentation algorithms.

Challenge homepage: https://amos22.grand-challenge.org/

Number of CT volumes: 200

Contrast: Contrast and non-contrast

CT body coverage: Abdomen

Does the dataset include any ground truth annotations?: Yes

Original GT annotation targets: (15 abdominal organs) spleen, right kidney, left kidney, gallbladder, esophagus, liver, stomach, aorta, inferior vena cava, pancreas, right adrenal gland, left adrenal gland, duodenum, bladder, prostate/uterus

Number of annotated CT volumes: 200

Annotator: AI + human refinement

Acquisition centers: Longgang District Central Hospital (SZ, CHINA) and Longgang District People's Hospital (SZ, CHINA).

Pathology/Disease: Patients diagnosed with abdominal tumors or other abnormalities; normal abdomen cases excluded

Original dataset download link: https://zenodo.org/records/7262581

Original dataset format: nifti