# VISCERAL Gold Corpus

<span style="background-color: #FFFF00"> Since the original host page (https://visceral.eu/benchmarks/anatomy3-open/) is no longer accessible, we have obtained renewed permission from the original authors and affiliated institutions to make the **VISCERAL** dataset available again as part of **CADS**, including its original images and annotations. </span>

The original images are provided under controlled access for academic, non-commercial research. Segmentations and other secondary data products are openly available here, while access to the original images requires an application.

👉 Apply for access via [this form](https://forms.gle/yXpGDCeFcgLeSqXM6)

## License
- Academic research use only
- Non-commercial use only
- Internal use only (not transferable, no sublicensing, no redistribution of original images)
- Access subject to the Data Access Agreement (DAA) confirmed in the application form

## Citation
Paper BibTeX:

```bibtex
@article{jimenez2016cloud,
  title={Cloud-based evaluation of anatomical structure segmentation and landmark detection algorithms: VISCERAL anatomy benchmarks},
  author={Jimenez-del-Toro, Oscar and M{\"u}ller, Henning and Krenn, Markus and Gruenberg, Katharina and Taha, Abdel Aziz and Winterstein, Marianne and Eggel, Ivan and Foncubierta-Rodr{\'\i}guez, Antonio and Goksel, Orcun and Jakab, Andr{\'a}s and others},
  journal={IEEE transactions on medical imaging},
  volume={35},
  number={11},
  pages={2459--2475},
  year={2016},
  publisher={IEEE}
}
```
```bibtex
@article{xu2025cads,
  title={CADS: A Comprehensive Anatomical Dataset and Segmentation for Whole-Body Anatomy in Computed Tomography},
  author={Xu, Murong and Amiranashvili, Tamaz and Navarro, Fernando and Fritsak, Maksym and Hamamci, Ibrahim Ethem and Shit, Suprosanna and Wittmann, Bastian and Er, Sezgin and Christ, Sebastian M. and de la Rosa, Ezequiel and Deseoe, Julian and Graf, Robert and Möller, Hendrik and Sekuboyina, Anjany and Peeken, Jan C. and Becker, Sven and Baldini, Giulia and Haubold, Johannes and Nensa, Felix and Hosch, René and Mirajkar, Nikhil and Khalid, Saad and Zachow, Stefan and Weber, Marc-André and Langs, Georg and Wasserthal, Jakob and Ozdemir, Mehmet Kemal and Fedorov, Andrey and Kikinis, Ron and Tanadini-Lang, Stephanie and Kirschke, Jan S. and Combs, Stephanie E. and Menze, Bjoern},
  journal={arXiv preprint arXiv:2507.22953},
  year={2025}
}
```

## Dataset description
The VISCERAL Anatomy benchmarks provide a cloud-based framework for evaluating algorithms on anatomical structure segmentation and landmark detection. The dataset includes 120 CT and MR patient volumes, with 1,295 structures and 1,760 landmarks manually annotated to form a standard Gold Corpus, supporting objective comparison of state-of-the-art methods.

**Homepage**: https://visceral.eu/benchmarks/anatomy3-open/ (currently down)

**Number of CT volumes**: 40

**Contrast**: Unenhanced whole-body CT and contrast-enhanced abdomen and thorax CT (iodine-based contrast agent)

**CT body coverage**: Whole-body (head to knee) and contrast-enhanced regions (corpus mandibulae to pelvis)

**Does the dataset include any ground truth annotations?**: Yes

**Original GT annotation targets**:

From VISCERAL Gold Corpus: Liver, spleen, pancreas, gallbladder, urinary bladder, aorta, trachea, right lung, left lung, sternum, thyroid gland, first lumbar vertebrae, right kidney, left kidney, right adrenal gland, left adrenal gland, right psoas major, left psoas major, right rectus abdominis, left rectus abdominis

From CADS's new manual annotation: Spinal canal, larynx, heart, mammary glands, colostomy bag, sigmoid colon, rectum, prostate, and seminal vesicles

**Number of annotated CT volumes**: 40

**Annotator**: Human

**Acquisition centers**: Medizinische Universität Wien (MUW), Universitätsklinikum Heidelberg (UKL-HD), Agència d‟Avalució, I Qualitat en Salut (GENCAT) in Catalonia Spain

**Pathology/Disease**: Whole-body: bone marrow neoplasms (e.g., multiple myeloma). Contrast-enhanced: malignant lymphoma. 

**Original dataset download link**: -

**Original dataset format**: nifti