CADS: a comprehensive anatomical dataset and segmentation for whole-body anatomy in computed tomography
Abstract Volumetric computed tomography (CT) segmentation supports quantitative image analysis, but most AI-based methods remain organ-specific, with fragmented models, inconsistent label definitions, and heterogeneous training and evaluation protocols. Here we present CADS, a unified data–model framework for scalable and generalizable whole-body CT segmentation. We construct a SNOMED-CT-referenced anatomical label space harmonizing heterogeneous cohorts from over 100 imaging centers, yielding complete annotations of 167 structures across 22,022 CT volumes and markedly expanding anatomical coverage and scale beyond existing resources. To complete labels on sparsely annotated CTs, we combine self-training with shape-informed quality control based on neural implicit functions, curating high-quality pseudo-labels. Trained on this resource, CADS shows robust whole-body segmentation across distribution shifts and improved performance relative to state-of-the-art approaches across 18 public datasets. We further validate it in an independent, expert-reviewed real-world oncology cohort of 2864 CTs. The curated dataset, pretrained models, and user-friendly software together provide an accessible, reproducible foundation for large-scale anatomical segmentation and computational anatomy research.
Authors
- Bjoern H. Menze (ORCID: https://orcid.org/0000-0003-4136-5690)
- Jan Stefan Kirschke (ORCID: https://orcid.org/0000-0002-7557-0003)
- Saad Khalid (ORCID: https://orcid.org/0009-0007-3371-1561)
- Stephanie Tanadini‐Lang (ORCID: https://orcid.org/0000-0002-4387-1522)
- Julian Deseoe (ORCID: https://orcid.org/0009-0003-6615-3851)
- Ron Kikinis (ORCID: https://orcid.org/0000-0001-7227-7058)
- Nikhil Mirajkar
- Suprosanna Shit (ORCID: https://orcid.org/0000-0003-4435-7207)
- René Hosch (ORCID: https://orcid.org/0000-0003-1760-2342)
- Tamaz Amiranashvili (ORCID: https://orcid.org/0000-0001-8914-3427)
- Andriy Fedorov (ORCID: https://orcid.org/0000-0003-4806-9413)
- Marc‐André Weber (ORCID: https://orcid.org/0000-0003-3918-8066)
- Robert Graf (ORCID: https://orcid.org/0000-0001-6656-3680)
- Sebastian M. Christ (ORCID: https://orcid.org/0000-0001-5330-3820)
- Sezgin Er (ORCID: https://orcid.org/0000-0001-7266-9844)
- Georg Langs (ORCID: https://orcid.org/0000-0002-5536-6873)
- İbrahim Ethem Hamamcı (ORCID: https://orcid.org/0000-0003-2932-3105)
- Giulia Baldini (ORCID: https://orcid.org/0000-0002-5929-0271)
- Sven Koitka (ORCID: https://orcid.org/0000-0001-9704-1180)
- Jan Caspar Peeken (ORCID: https://orcid.org/0000-0003-2679-9853)
- Hendrik Möller (ORCID: https://orcid.org/0009-0001-1978-5894)
- Ezequiel de la Rosa (ORCID: https://orcid.org/0000-0002-9042-1962)
- Jakob Wasserthal (ORCID: https://orcid.org/0000-0002-9921-5698)
- Bastian Wittmann (ORCID: https://orcid.org/0000-0002-6308-2926)
- Stefan Zachow A (ORCID: https://orcid.org/0000-0001-7964-3049)
- Felix Nensa (ORCID: https://orcid.org/0000-0002-5811-7100)
- Murong Xu (ORCID: https://orcid.org/0009-0008-4220-4403)
- Maksym Fritsak (ORCID: https://orcid.org/0009-0008-9443-0082)
- Anjany Sekuboyina
- Johannes Haubold
- Stephanie E. Combs
- Mehmet Kemal Ozdemir
- Fernando Navarro
Publication Details
- Journal
- Nature Communications
- Published
- 2026-10-28
- DOI
- https://doi.org/10.1038/s41467-026-78732-1
- Primary Topic
- Medical Image Segmentation Techniques
- Type
- article
- Field-Weighted Citation Impact
- 0.00