BiFedKD: Bidirectional Federated Knowledge Distillation Framework for Non-IID and Long-Tailed ECG Monitoring

2026-05-14

AI Summary

This study addresses key challenges in ECG monitoring within the Internet of Medical Things, including data privacy concerns, limited communication bandwidth, and degraded federated learning performance caused by non-IID and long-tailed label distributions. To tackle these issues, the authors propose BiFedKD, a bidirectional federated knowledge distillation framework that innovatively integrates bidirectional distillation aggregation with a temperature scaling mechanism to generate stable global logit signals. This approach effectively enhances model alignment and generalization across clients. Evaluated on the MIT-BIH Arrhythmia dataset, BiFedKD achieves a 3.52% improvement in accuracy and a 9.93% gain in Macro-F1 score over baseline methods, while reducing communication overhead by 40% and computational cost by 71.7% to reach comparable performance.

Abstract

Electrocardiogram (ECG) monitoring in Internet of Medical Things (IoMT) networks is constrained by strict data-sharing regulations and privacy concerns. Federated learning (FL) enables collaborative learning by keeping raw ECG data on devices, but frequent transmissions of high-dimensional model updates incur heavy per-round traffic over bandwidth-limited links. To alleviate this bottleneck, federated distillation (FD) replaces parameter exchange with logit-based knowledge transfer. However, the performance of FD often degrades under the non-independent and identically distributed (non-IID) and long-tailed label distributions in ECG deployments. To address these challenges, we propose a bidirectional federated knowledge distillation (BiFedKD) framework that employs an aggregation-by-distillation pipeline with temperature scaling to produce a stable global distillation signal for cross-client alignment. Experiments on the MIT-BIH Arrhythmia dataset show that BiFedKD improves accuracy and Macro-F1 over the baseline by $3.52\%$ and $9.93\%$, respectively. Moreover, to reach the same Macro-F1, BiFedKD reduces communication overhead by $40\%$ and computation cost by $71.7\%$ compared with the baseline.

Problem

• Federated Learning
• Non-IID
• Long-Tailed Distribution
• ECG Monitoring
• Communication Overhead

Innovation

• Bidirectional Federated Knowledge Distillation
• Non-IID
• Long-Tailed Distribution
• Temperature Scaling
• Communication Efficiency