Evaluating Self-Supervised Learning in Medical Imaging: A Systematic Investigation of Robustness, Generalizability, and Multi-Domain Impact


Valay Bundele (University of Tuebingen), Karahan Sarıtaş (University of Tuebingen), Bora Kargi (University of Tuebingen), Oğuz Ata Çal (University of Tuebingen), Kıvanç Tezören (University of Tuebingen), Zohreh Ghaderi (University of Tuebingen), Hendrik Lensch (University of Tuebingen)
The 35th British Machine Vision Conference
PDFPosterVideo (Right click to download)Supplementary

Abstract

Self-supervised learning (SSL) has emerged as a promising paradigm in medical imaging, addressing the chronic challenge of limited labeled data in healthcare settings. While SSL has shown impressive results, existing studies in the medical domain are often limited in scope, focusing on specific datasets or modalities, or evaluating only isolated aspects of model performance. This fragmented evaluation approach poses a significant challenge, as models deployed in critical medical settings must not only achieve high accuracy but also demonstrate robust performance and generalizability across diverse datasets and varying conditions. To bridge this gap, we conduct a rigorous investigation into the design space of SSL for medical imaging, evaluating 8 major SSL methods across 11 real-world medical datasets. Our analysis spans three core dimensions: (1) in-domain performance under varying label proportions (1\%, 10\%, and 100\%), (2) cross-dataset generalization, and (3) robustness to out-of-distribution (OOD) samples. Beyond empirical evaluation, we further examine how initialization strategies, model architectures, and multi-domain pre-training contribute to SSL’s success in medical imaging.

Citation

@inproceedings{Bundele_2025_BMVC,
author    = {Valay Bundele and Karahan Sarıtaş and Bora Kargi and Oğuz Ata Çal and Kıvanç Tezören and Zohreh Ghaderi and Hendrik Lensch},
title     = {Evaluating Self-Supervised Learning in Medical Imaging:  A Systematic Investigation of Robustness, Generalizability, and Multi-Domain Impact},
booktitle = {36th British Machine Vision Conference 2025, {BMVC} 2025, Sheffield, UK, November 24-27, 2025},
publisher = {BMVA},
year      = {2025},
url       = {https://bmva-archive.org.uk/bmvc/2025/assets/papers/Paper_1123/paper.pdf}
}


Copyright © 2025 The British Machine Vision Association and Society for Pattern Recognition
The British Machine Vision Conference is organised by The British Machine Vision Association and Society for Pattern Recognition. The Association is a Company limited by guarantee, No.2543446, and a non-profit-making body, registered in England and Wales as Charity No.1002307 (Registered Office: Dept. of Computer Science, Durham University, South Road, Durham, DH1 3LE, UK).

Imprint | Data Protection