CFL-SparseMed: Communication-Efficient Federated Learning for Medical Imaging with Top-k Sparse Updates

2025-10-25

AI Summary

To address data silos, non-independent and identically distributed (Non-IID) data, and high communication overhead in medical image classification, this paper proposes a privacy-preserving sparse federated learning framework. Methodologically, it introduces a Top-k gradient sparsification mechanism, integrates adaptive federated averaging with a Non-IID-aware local model optimization strategy, and incorporates differential privacy and secure aggregation for end-to-end privacy protection. Empirically evaluated on multi-institutional medical imaging benchmarks—including CheXpert and BraTS—the framework reduces communication costs by up to 90% while maintaining or even improving model accuracy. Results demonstrate robustness and strong generalization across heterogeneous clinical datasets. The proposed approach jointly optimizes training efficiency, model performance, and privacy compliance, offering a scalable and regulation-adherent collaborative paradigm for clinical-grade distributed intelligent diagnosis.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Computer Vision: Multi-modal Vision
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Security and Privacy: Data transparency and provenance
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing

Abstract

Secure and reliable medical image classification is crucial for effective patient treatment, but centralized models face challenges due to data and privacy concerns. Federated Learning (FL) enables privacy-preserving collaborations but struggles with heterogeneous, non-IID data and high communication costs, especially in large networks. We propose extbf{CFL-SparseMed}, an FL approach that uses Top-k Sparsification to reduce communication overhead by transmitting only the top k gradients. This unified solution effectively addresses data heterogeneity while maintaining model accuracy. It enhances FL efficiency, preserves privacy, and improves diagnostic accuracy and patient care in non-IID medical imaging settings. The reproducibility source code is available on href{https://github.com/Aniket2241/APK_contruct}{Github}.

Problem

• Reduces communication costs in federated medical imaging
• Addresses data heterogeneity in non-IID medical datasets
• Maintains model accuracy while preserving patient privacy

Innovation

• Top-k sparsification reduces communication overhead
• Unified solution addresses data heterogeneity
• Maintains model accuracy while preserving privacy