Is Self-supervision Enough?: Benchmarking Foundation Models Against End-to-end Training for Mitotic Figure Classification

Ganz J, Ammeling J, Rosbach E, Lausser L, Bertram CA, Breininger K, Aubreville M (2025)

Publication Type: Conference contribution

Publication year: 2025

Journal

Informatik aktuell Gesellschaft für Informatik (GI)

Publisher: Springer Science and Business Media Deutschland GmbH

Pages Range: 63-68

Conference Proceedings Title: Informatik aktuell

Event location: Regensburg, DEU

ISBN: 9783658474218

DOI: 10.1007/978-3-658-47422-5_15

Abstract

Foundation models (FMs), i.e., models trained on a vast amount of typically unlabeled data, have become popular and available recently for the domain of histopathology. The key idea is to extract semantically rich vectors from any input patch, allowing for the use of simple subsequent classification networks potentially reducing the required amounts of labeled data, and increasing domain robustness. In this work, we investigate to which degree this also holds for mitotic figure classification. Utilizing two popular public mitotic figure datasets, we compared linear probing of five publicly available FMs against models trained on ImageNet and a simple ResNet50 end-to-end-trained baseline. We found that the end-to-end-trained baseline outperformed all FM-based classifiers, regardless of the amount of data provided. Additionally, we did not observe the FM-based classifiers to be more robust against domain shifts, rendering both of the above assumptions incorrect.

Additional Organisation(s)

Sonderforschungsbereich 1540/1 - Erforschung der Mechanik des Gehirns (EBM): Verständnis, Engineering und Nutzung mechanischer Eigenschaften und Signale in der Entwicklung, Physiologie und Pathologie des zentralen Nervensystems

Involved external institutions

Technische Hochschule Ingolstadt DE Germany (DE) Veterinärmedizinische Universität Wien (Vetmeduni Vienna) / University of Veterinary Medicine, Vienna AT Austria (AT) Julius-Maximilians-Universität Würzburg DE Germany (DE) Fachhochschule Flensburg DE Germany (DE)

How to cite

APA:

Ganz, J., Ammeling, J., Rosbach, E., Lausser, L., Bertram, C.A., Breininger, K., & Aubreville, M. (2025). Is Self-supervision Enough?: Benchmarking Foundation Models Against End-to-end Training for Mitotic Figure Classification. In Christoph Palm, Katharina Breininger, Thomas Deserno, Heinz Handels, Andreas Maier, Klaus H. Maier-Hein, Thomas M. Tolxdorff (Eds.), Informatik aktuell (pp. 63-68). Regensburg, DEU: Springer Science and Business Media Deutschland GmbH.

MLA:

Ganz, Jonathan, et al. "Is Self-supervision Enough?: Benchmarking Foundation Models Against End-to-end Training for Mitotic Figure Classification." Proceedings of the German Conference on Medical Image Computing, 2025, Regensburg, DEU Ed. Christoph Palm, Katharina Breininger, Thomas Deserno, Heinz Handels, Andreas Maier, Klaus H. Maier-Hein, Thomas M. Tolxdorff, Springer Science and Business Media Deutschland GmbH, 2025. 63-68.

BibTeX: Download