Standard
Appearance-based Debiasing of Deep Learning Models in Medical Imaging. / Wilm, Frauke; Reimann, Marcel; Taubmann, Oliver; Mühlberg, Alexander; Breininger, Katharina.
Bildverarbeitung für die Medizin 2024 : Proceedings, German Conference on Medical Image Computing, 2024. ed. / Andreas Maier; Thomas M. Deserno; Heinz Handels; Klaus Maier-Hein; Christoph Palm; Thomas Tolxdorff. Springer Science and Business Media Deutschland GmbH, 2024. p. 19-24 (Informatik aktuell).
Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Harvard
Wilm, F
, Reimann, M, Taubmann, O, Mühlberg, A & Breininger, K 2024,
Appearance-based Debiasing of Deep Learning Models in Medical Imaging. in A Maier, TM Deserno, H Handels, K Maier-Hein, C Palm & T Tolxdorff (eds),
Bildverarbeitung für die Medizin 2024 : Proceedings, German Conference on Medical Image Computing, 2024. Springer Science and Business Media Deutschland GmbH, Informatik aktuell, pp. 19-24, German Conference on Medical Image Computing, BVM 2024, Erlangen, Germany,
10/03/2024.
https://doi.org/10.1007/978-3-658-44037-4_9
APA
Wilm, F.
, Reimann, M., Taubmann, O., Mühlberg, A., & Breininger, K. (2024).
Appearance-based Debiasing of Deep Learning Models in Medical Imaging. In A. Maier, T. M. Deserno, H. Handels, K. Maier-Hein, C. Palm, & T. Tolxdorff (Eds.),
Bildverarbeitung für die Medizin 2024 : Proceedings, German Conference on Medical Image Computing, 2024 (pp. 19-24). Springer Science and Business Media Deutschland GmbH. Informatik aktuell
https://doi.org/10.1007/978-3-658-44037-4_9
Vancouver
Wilm F
, Reimann M, Taubmann O, Mühlberg A, Breininger K.
Appearance-based Debiasing of Deep Learning Models in Medical Imaging. In Maier A, Deserno TM, Handels H, Maier-Hein K, Palm C, Tolxdorff T, editors, Bildverarbeitung für die Medizin 2024 : Proceedings, German Conference on Medical Image Computing, 2024. Springer Science and Business Media Deutschland GmbH. 2024. p. 19-24. (Informatik aktuell).
https://doi.org/10.1007/978-3-658-44037-4_9
Author
Wilm, Frauke ; Reimann, Marcel ; Taubmann, Oliver ; Mühlberg, Alexander ; Breininger, Katharina. / Appearance-based Debiasing of Deep Learning Models in Medical Imaging. Bildverarbeitung für die Medizin 2024 : Proceedings, German Conference on Medical Image Computing, 2024. editor / Andreas Maier ; Thomas M. Deserno ; Heinz Handels ; Klaus Maier-Hein ; Christoph Palm ; Thomas Tolxdorff. Springer Science and Business Media Deutschland GmbH, 2024. pp. 19-24 (Informatik aktuell).
Bibtex
@inproceedings{781415d31e074aa8bd74f12facfbe220,
title = "Appearance-based Debiasing of Deep Learning Models in Medical Imaging",
abstract = "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.",
author = "Frauke Wilm and Marcel Reimann and Oliver Taubmann and Alexander M{\"u}hlberg and Katharina Breininger",
note = "Publisher Copyright: {\textcopyright} Der/die Autor(en), exklusiv lizenziert an Springer Fachmedien Wiesbaden GmbH, ein Teil von Springer Nature 2024.; German Conference on Medical Image Computing, BVM 2024 ; Conference date: 10-03-2024 Through 12-03-2024",
year = "2024",
doi = "10.1007/978-3-658-44037-4_9",
language = "English",
isbn = "9783658440367",
series = "Informatik aktuell",
pages = "19--24",
editor = "Andreas Maier and Deserno, {Thomas M.} and Heinz Handels and Klaus Maier-Hein and Christoph Palm and Thomas Tolxdorff",
booktitle = "Bildverarbeitung f{\"u}r die Medizin 2024",
publisher = "Springer Science and Business Media Deutschland GmbH",
address = "Germany",
}
RIS
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 -