Asymmetrical Reciprocity-based Federated Learning for Resolving Disparities in Medical Diagnosis

TL;DR

This paper tackles geographic health disparities by introducing FedHelp, a cross-silo federated learning framework that serves underserved regions with small data through one-time API access to foundation models. It addresses asymmetrical reciprocity between small and large clients via an asymmetric dual knowledge distillation mechanism, enabling effective knowledge transfer without data sharing. Empirical results on melanoma and pneumonia classification and on 2D/3D segmentation demonstrate strong gains for small clients and robust performance across tasks, with notable reductions in communication cost when using proxy models. The approach offers a scalable path to improving diagnostic accuracy in underserved settings and broader applicability to medical imaging tasks that involve heterogeneous client capacities.

Abstract

Geographic health disparities pose a pressing global challenge, particularly in underserved regions of low- and middle-income nations. Addressing this issue requires a collaborative approach to enhance healthcare quality, leveraging support from medically more developed areas. Federated learning emerges as a promising tool for this purpose. However, the scarcity of medical data and limited computation resources in underserved regions make collaborative training of powerful machine learning models challenging. Furthermore, there exists an asymmetrical reciprocity between underserved and developed regions. To overcome these challenges, we propose a novel cross-silo federated learning framework, named FedHelp, aimed at alleviating geographic health disparities and fortifying the diagnostic capabilities of underserved regions. Specifically, FedHelp leverages foundational model knowledge via one-time API access to guide the learning process of underserved small clients, addressing the challenge of insufficient data. Additionally, we introduce a novel asymmetric dual knowledge distillation module to manage the issue of asymmetric reciprocity, facilitating the exchange of necessary knowledge between developed large clients and underserved small clients. We validate the effectiveness and utility of FedHelp through extensive experiments on both medical image classification and segmentation tasks. The experimental results demonstrate significant performance improvement compared to state-of-the-art baselines, particularly benefiting clients in underserved regions.

Figures (5)

• Figure 1: Client accuracy comparison between client-wise local training, FedAvg, and the proposed FedHelp on the Fed-ISIC19 dataset. The size of each client can be found in Section \ref{['sec:melanoma']}.
• Figure 2: Overview of the proposed FedHelp framework. "CE"/"KL" denotes the cross-entropy loss/Kullback–Leibler divergence.
• Figure 3: Averge client accuracy of two ablation studies on melanoma classification
• Figure 4: Visualization of 2D and 3D segmentation tasks.
• Figure 5: Abalation study results on segmentation task.