FairFedMed: Achieving Equity in Medical Federated Learning via FairLoRA

Minghan Li, Congcong Wen, Yu Tian, Min Shi, Yan Luo, Hao Huang, Yi Fang, Mengyu Wang

Keywords: Fairness, Equity, Federated Learning, FairLoRA

Abstract: Fairness remains a critical concern in healthcare, where unequal access to services and treatment outcomes can adversely affect patient health. While Federated Learning (FL) presents a collaborative and privacy-preserving approach to model training, ensuring fairness is challenging due to heterogeneous data across institutions, and current research primarily addresses non-medical applications. To fill this gap, we introduce FairFedMed, the first FL dataset specifically designed to study group fairness (i.e., demographics) in the medical field. It consists of paired 2D SLO funfus images and 3D OCT B-Scans from 15,165 glaucoma patients, along with six different demographic attributes. Existing state-of-the-art FL models may work well for natural images but often struggle with medical images due to their unique characteristics. Moreover, these models do not sufficiently address performance disparities across diverse demographic groups. To overcome these limitations, we propose FairLoRA, a novel fairness-aware FL framework based on singular value decomposition(SVD)-based low-rank approximation. FairLoRA incorporates customized singular value matrices for each demographic group and shares singular vector matrices across all demographic groups, ensuring both model equity and computational efficiency. Experimental results on the FairFedMed dataset demonstrate that FairLoRA not only achieves state-of-the-art performance in medical image classification but also significantly improves fairness across diverse populations. Our code and dataset can be accessible via the Github anonymous link: https://github.com/Anonymouse4Science/FairFedMed-FairLoRA.git

Primary Area: alignment, fairness, safety, privacy, and societal considerations

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 2636