Shortcut learning leads to sex bias in deep learning models for photoacoustic tomography
Shortcut learning has been identified as a source of algorithmic unfairness in medical imaging artificial intelligence (AI), but its impact on photoacoustic tomography (PAT), particularly concerning sex bias, remains underexplored. This study investigates this issue using peripheral artery disease (...
Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Article (Journal) |
| Language: | English |
| Published: |
July 2025
|
| In: |
International journal of computer assisted radiology and surgery
Year: 2025, Volume: 20, Issue: 7, Pages: 1325-1333 |
| ISSN: | 1861-6429 |
| DOI: | 10.1007/s11548-025-03370-9 |
| Online Access: | Verlag, lizenzpflichtig, Volltext: https://doi.org/10.1007/s11548-025-03370-9 Verlag, lizenzpflichtig, Volltext: https://link.springer.com/article/10.1007/s11548-025-03370-9 |
| Author Notes: | Marcel Knopp, Christoph J. Bender, Niklas Holzwarth, Yi Li, Julius Kempf, Milenko Caranovic, Ferdinand Knieling, Werner Lang, Ulrich Rother, Alexander Seitel, Lena Maier-Hein, Kris K. Dreher |
| Summary: | Shortcut learning has been identified as a source of algorithmic unfairness in medical imaging artificial intelligence (AI), but its impact on photoacoustic tomography (PAT), particularly concerning sex bias, remains underexplored. This study investigates this issue using peripheral artery disease (PAD) diagnosis as a specific clinical application. |
|---|---|
| Item Description: | Online verfügbar: 09. Mai 2025 Gesehen am 23.10.2025 |
| Physical Description: | Online Resource |
| ISSN: | 1861-6429 |
| DOI: | 10.1007/s11548-025-03370-9 |