TY - GEN

T1 - Appearance-based Debiasing of Deep Learning Models in Medical Imaging

AU - Wilm, Frauke

AU - Reimann, Marcel

AU - Taubmann, Oliver

AU - Mühlberg, Alexander

AU - Breininger, Katharina

N1 - Publisher Copyright: © Der/die Autor(en), exklusiv lizenziert an Springer Fachmedien Wiesbaden GmbH, ein Teil von Springer Nature 2024.

PY - 2024

Y1 - 2024

N2 - Out-of-distribution data can substantially impede the performance of deep learning models. In medical imaging, domain shifts can, for instance, be caused by different image acquisition protocols. To address these domain shifts, domain adversarial training can be employed to constrain a model to domainagnostic features. This, however, requires prior knowledge about the domain variable, which might not always be accessible. Recent approaches make use of control regions to guide the training process and thereby alleviate the need for prior domain knowledge. In this work, we combine these approaches with traditional domain adversarial training to exploit the benefits of both methods.We test the proposed method on two medical datasets and demonstrate performance increases of up to 10 %, compared to a baseline trained without debiasing.

AB - Out-of-distribution data can substantially impede the performance of deep learning models. In medical imaging, domain shifts can, for instance, be caused by different image acquisition protocols. To address these domain shifts, domain adversarial training can be employed to constrain a model to domainagnostic features. This, however, requires prior knowledge about the domain variable, which might not always be accessible. Recent approaches make use of control regions to guide the training process and thereby alleviate the need for prior domain knowledge. In this work, we combine these approaches with traditional domain adversarial training to exploit the benefits of both methods.We test the proposed method on two medical datasets and demonstrate performance increases of up to 10 %, compared to a baseline trained without debiasing.

U2 - 10.1007/978-3-658-44037-4_9

DO - 10.1007/978-3-658-44037-4_9

M3 - Article in proceedings

AN - SCOPUS:85188267480

SN - 9783658440367

T3 - Informatik aktuell

SP - 19

EP - 24

BT - Bildverarbeitung für die Medizin 2024

A2 - Maier, Andreas

A2 - Deserno, Thomas M.

A2 - Handels, Heinz

A2 - Maier-Hein, Klaus

A2 - Palm, Christoph

A2 - Tolxdorff, Thomas

PB - Springer Science and Business Media Deutschland GmbH

T2 - German Conference on Medical Image Computing, BVM 2024

Y2 - 10 March 2024 through 12 March 2024

ER -