FCA: Taming Long-tailed Federated Medical Image Classification by Classifier Anchoring

TL;DR

This paper introduces federated classifier anchoring (FCA), a novel method that improves federated learning for long-tailed medical image classification by debiasing classifiers and enhancing both global and local model performance.

Contribution

FCA adds personalized classifiers at each client and employs consistency learning to mitigate class imbalance and divergence in federated medical image classification.

Findings

FCA outperforms state-of-the-art methods in skin lesion classification.

FCA improves model generalization and specialization across clients.

FCA effectively handles class imbalance in federated learning.

Abstract

Limited training data and severe class imbalance impose significant challenges to developing clinically robust deep learning models. Federated learning (FL) addresses the former by enabling different medical clients to collaboratively train a deep model without sharing data. However, the class imbalance problem persists due to inter-client class distribution variations. To overcome this, we propose federated classifier anchoring (FCA) by adding a personalized classifier at each client to guide and debias the federated model through consistency learning. Additionally, FCA debiases the federated classifier and each client's personalized classifier based on their respective class distributions, thus mitigating divergence. With FCA, the federated feature extractor effectively learns discriminative features suitably globally for federation as well as locally for all participants. In clinical practice, the federated model is expected to be both generalized, performing well across clients, and specialized, benefiting each individual client from collaboration. According to this, we propose a novel evaluation metric to assess models' generalization and specialization performance globally on an aggregated public test set and locally at each client. Through comprehensive comparison and evaluation, FCA outperforms the state-of-the-art methods with large margins for federated long-tailed skin lesion classification and intracranial hemorrhage classification, making it a more feasible solution in clinical settings.