Automated AI detection of thoracic aortic dissection on CT imaging
Early detection of aortic dissection (AD) is crucial for timely treatment. A modern convolutional neural network (CNN) achieved 93.5% sensitivity and 100.0% specificity for AD detection on multicenter, heterogeneous CT data. These results demonstrate the potential of streamlined, optimized CNNs for robust AD detection on CT, supporting fast clinical response.
Authors
- Florian Haag (ORCID: https://orcid.org/0009-0008-8167-0503)
- Jesus G. Ulloa (ORCID: https://orcid.org/0000-0002-0311-1633)
- Nils Rathmann (ORCID: https://orcid.org/0000-0002-2795-5611)
- Caterina Rizzo (ORCID: https://orcid.org/0000-0003-1299-1756)
- Alexander Hertel (ORCID: https://orcid.org/0000-0002-5438-1887)
- Lucas R. Cusumano (ORCID: https://orcid.org/0000-0001-9710-8488)
- Johann S. Rink (ORCID: https://orcid.org/0000-0002-7069-5181)
- Tobias Norajitra
- Michael Baumgartner (ORCID: https://orcid.org/0000-0003-4455-9917)
- Stefan Oswald Schoenberg (ORCID: https://orcid.org/0000-0002-3115-8367)
- Klaus H. Maier-Hein
- Steffen J. Diehl
Institutions
- German Cancer Research Center (DE)
- Heidelberg University (DE)
- Ronald Reagan UCLA Medical Center (US)
- University Hospital Heidelberg (DE)
- Hospital Alemán (AR)
- University Medical Centre Mannheim (DE)
- Hospital General de Agudos Dr. TEODORO ALVAREZ (AR)
Publication Details
- Journal
- European Radiology Experimental
- Published
- 2025-10-22
- DOI
- https://doi.org/10.1186/s41747-025-00640-8
- Citations
- 1
- Primary Topic
- Aortic Disease and Treatment Approaches
- Type
- article
- Field-Weighted Citation Impact
- 1.52