CT-ORG: Multiple Organ Segmentation in CT

License

CC BY 3.0
Creative Commons Attribution 3.0 License

Citation

Paper BibTeX:

@article{rister2020ct,
  title={CT-ORG, a new dataset for multiple organ segmentation in computed tomography},
  author={Rister, Blaine and Yi, Darvin and Shivakumar, Kaushik and Nobashi, Tomomi and Rubin, Daniel L},
  journal={Scientific Data},
  volume={7},
  number={1},
  pages={381},
  year={2020},
  publisher={Nature Publishing Group UK London}
}

Dataset:

Rister, B., Shivakumar, K., Nobashi, T., & Rubin, D. L. (2019). CT-ORG: A Dataset of CT Volumes With Multiple Organ Segmentations (Version 1) [dataset]. The Cancer Imaging Archive.  DOI: 10.7937/tcia.2019.tt7f4v7o

Dataset description

CT-ORG contains 140 CT scans from diverse sources, each with 3D segmentations of five organs, and brain labels in some cases. The dataset covers a wide range of imaging conditions and includes both benign and malignant liver lesions, as well as metastatic disease in bones and lungs, providing a challenging benchmark for multi-class organ segmentation.

Number of CT volumes: 140

Contrast: Both contrast-enhanced and non-contrast; includes PET-CT derived scans

CT body coverage: Abdominal and full-body

Does the dataset include any ground truth annotations?: Yes

Original GT annotation targets: Liver, urinary bladder, lungs, kidneys, bone

Number of annotated CT volumes: 140

Annotator: Human (lungs and bones partly from morphological algorithms)

Acquisition centers: Multiple global institutions, Ludwig Maxmilian University of Munich, Radboud University Medical Center of Nijmegen, Poly-technique & CHUM Research Center Montreal, Tel Aviv University, Sheba Medical Center, IRCAD Institute Strasbourg and Hebrew University of Jerusalem. The PET-CT images all derive from Stanford Healthcare.

Pathology/Disease: Benign and malignant liver lesions, metastatic disease in bones and lungs

Original dataset download link: https://www.cancerimagingarchive.net/collection/ct-org/

Original dataset format: nifti