TY  - JOUR
AU  - Knopp, Marcel
AU  - Bender, Christoph Julien
AU  - Holzwarth, Niklas
AU  - Li, Yi
AU  - Kempf, Julius
AU  - Caranovic, Milenko
AU  - Knieling, Ferdinand
AU  - Lang, Werner
AU  - Rother, Ulrich
AU  - Seitel, Alexander
AU  - Maier-Hein, Lena
AU  - Dreher, Kris
TI  - Shortcut learning leads to sex bias in deep learning models for photoacoustic tomography.
JO  - International journal of computer assisted radiology and surgery
VL  - 20
IS  - 7
SN  - 1861-6410
CY  - Heidelberg [u.a.]
PB  - Springer
M1  - DKFZ-2025-00955
SP  - 1325-1333
PY  - 2025
N1  - #EA:E130#LA:E130# / 2025 Jul;20(7):1325-1333
AB  - 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.To examine the potential for sex bias due to shortcut learning in convolutional neural network (CNNs) and assess how such biases might affect diagnostic predictions, we created training and test datasets with varying PAD prevalence between sexes. Using these datasets, we explored (1) whether CNNs can classify the sex from imaging data, (2) how sex-specific prevalence shifts impact PAD diagnosis performance and underdiagnosis disparity between sexes, and (3) how similarly CNNs encode sex and PAD features.Our study with 147 individuals demonstrates that CNNs can classify the sex from calf muscle PAT images, achieving an AUROC of 0.75. For PAD diagnosis, models trained on data with imbalanced sex-specific disease prevalence experienced significant performance drops (up to 0.21 AUROC) when applied to balanced test sets. Additionally, greater imbalances in sex-specific prevalence within the training data exacerbated underdiagnosis disparities between sexes. Finally, we identify evidence of shortcut learning by demonstrating the effective reuse of learned feature representations between PAD diagnosis and sex classification tasks.CNN-based models trained on PAT data may engage in shortcut learning by leveraging sex-related features, leading to biased and unreliable diagnostic predictions. Addressing demographic-specific prevalence imbalances and preventing shortcut learning is critical for developing models in the medical field that are both accurate and equitable across diverse patient populations.
KW  - Peripheral artery disease (PAD) (Other)
KW  - Photoacoustic tomography (PAT) (Other)
KW  - Sex Bias in AI (Other)
KW  - Shortcut learning (Other)
LB  - PUB:(DE-HGF)16
C6  - pmid:40343639
DO  - DOI:10.1007/s11548-025-03370-9
UR  - https://inrepo02.dkfz.de/record/301270
ER  -