FedDAG: Federated Domain Adversarial Generation Towards Generalizable Medical Image Analysis

2025-01-22

AI Summary

To address weak domain generalization caused by client data heterogeneity in federated medical image analysis, this paper proposes a federated domain generalization framework tailored for unknown target domains. Methodologically: (1) an end-to-end adversarial domain generation module is designed to synthesize novel images with显著 style perturbations—distinct from all local source domains—to explicitly model cross-domain feature shifts; and (2) a hierarchical sharpness-aware aggregation mechanism is introduced for the first time in federated learning to mitigate client contribution imbalance under data heterogeneity. Evaluated on four medical imaging benchmarks, the method achieves average cross-domain accuracy gains of 4.2–7.8%, substantially improving model robustness and generalization to unseen target domains. Empirical results validate its effectiveness and practicality in real-world federated healthcare settings.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Adversarial Attacks & Robustness
• Application Domains: Misinformation & Fake News

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing

Abstract

Federated domain generalization aims to train a global model from multiple source domains and ensure its generalization ability to unseen target domains. {Due to the target domain being with unknown domain shifts, attempting to approximate these gaps by source domains may be the key to improving model generalization capability.} Existing works mainly focus on sharing and recombining local domain-specific attributes to increase data diversity and simulate potential domain shifts. {However, these methods may be insufficient since only the local attribute recombination can be hard to touch the out-of-distribution of global data.} In this paper, we propose a simple-yet-efficient framework named Federated Domain Adversarial Generation (FedDAG). {It aims to simulate the domain shift and improve the model generalization by adversarially generating novel domains different from local and global source domains.} Specifically, it generates novel-style images by maximizing the instance-level feature discrepancy between original and generated images and trains a generalizable task model by minimizing their feature discrepancy. {Further, we observed that FedDAG could cause different performance improvements for local models. It may be due to inherent data isolation and heterogeneity among clients, exacerbating the imbalance in their generalization contributions to the global model.} {Ignoring this imbalance can lead the global model's generalization ability to be sub-optimal, further limiting the novel domain generation procedure. } Thus, to mitigate this imbalance, FedDAG hierarchically aggregates local models at the within-client and across-client levels by using the sharpness concept to evaluate client model generalization contributions. {Extensive experiments across four medical benchmarks demonstrate FedDAG's ability to enhance generalization in federated medical scenarios.}

Problem

• Domain Generalization
• Medical Image Analysis
• Model Generalization

Innovation

• FedDAG
• Domain Generalization
• Federated Learning