Mind the Gap: Scanner-Induced Domain Shifts Pose Challenges for Representation Learning in Histopathology

Wilm F, Fragoso M, Bertram CA, Stathonikos N, Ottl M, Qiu J, Klopfleisch R, Maier A, Aubreville M, Breininger K (2023)

Publication Type: Conference contribution

Publication year: 2023

Journal

Proceedings - International Symposium on Biomedical Imaging

Publisher: IEEE Computer Society

Book Volume: 2023-April

Conference Proceedings Title: Proceedings - International Symposium on Biomedical Imaging

Event location: Cartagena CO

ISBN: 9781665473583

DOI: 10.1109/ISBI53787.2023.10230458

Abstract

Computer-aided systems in histopathology are often challenged by various sources of domain shift that impact the performance of these algorithms considerably. We investigated the potential of using self-supervised pre-training to overcome scanner-induced domain shifts for the downstream task of tumor segmentation. For this, we present the Barlow Triplets to learn scanner-invariant representations from a multi-scanner dataset with local image correspondences. We show that self-supervised pre-training successfully aligned different scanner representations, which, interestingly only results in a limited benefit for our downstream task. We thereby provide insights into the influence of scanner characteristics for downstream applications and contribute to a better understanding of why established self-supervised methods have not yet shown the same success on histopathology data as they have for natural images.

Authors with CRIS profile

Frauke Wilm Lehrstuhl für Informatik 5 (Mustererkennung) Jingna Qiu Juniorprofessur für Artificial Intelligence in Medical Imaging Andreas Maier Lehrstuhl für Informatik 5 (Mustererkennung) Katharina Breininger Juniorprofessur für Artificial Intelligence in Medical Imaging

Involved external institutions

Freie Universität Berlin DE Germany (DE) Veterinärmedizinische Universität Wien (Vetmeduni Vienna) / University of Veterinary Medicine, Vienna AT Austria (AT) University Medical Centre Utrecht (UMC Utrecht) NL Netherlands (NL) Technische Hochschule Ingolstadt DE Germany (DE)

How to cite

APA:

Wilm, F., Fragoso, M., Bertram, C.A., Stathonikos, N., Ottl, M., Qiu, J.,... Breininger, K. (2023). Mind the Gap: Scanner-Induced Domain Shifts Pose Challenges for Representation Learning in Histopathology. In Proceedings - International Symposium on Biomedical Imaging. Cartagena, CO: IEEE Computer Society.

MLA:

Wilm, Frauke, et al. "Mind the Gap: Scanner-Induced Domain Shifts Pose Challenges for Representation Learning in Histopathology." Proceedings of the 20th IEEE International Symposium on Biomedical Imaging, ISBI 2023, Cartagena IEEE Computer Society, 2023.

BibTeX: Download