Federated Learning with Gramian Angular Fields for Privacy-Preserving ECG Classification on Heterogeneous IoT Devices

TL;DR

This paper tackles privacy-preserving ECG classification in IoT environments by marrying Federated Learning with Gramian Angular Field representations, enabling CNN-based analysis on 2D ECG images while keeping data local to devices. The authors implement a server-plus-two-clients FL framework and validate it on heterogeneous hardware (server, laptop, Raspberry Pi), using 32×32 GAF images and a CNN with four convolutional layers and a 128-neuron dense layer for classification into five heartbeat types. They report a multi-client test accuracy of 95.18% with reduced training time compared to a single-client setting, while highlighting trade-offs in communication overhead and the impact of device heterogeneity and non-IID data. The study demonstrates the feasibility and practicality of scalable, privacy-preserving ECG analytics in edge–cloud IoT health systems and points to future enhancements in compression, personalization, and acceleration for real-time deployment.

Abstract

This study presents a federated learning (FL) framework for privacy-preserving electrocardiogram (ECG) classification in Internet of Things (IoT) healthcare environments. By transforming 1D ECG signals into 2D Gramian Angular Field (GAF) images, the proposed approach enables efficient feature extraction through Convolutional Neural Networks (CNNs) while ensuring that sensitive medical data remain local to each device. This work is among the first to experimentally validate GAF-based federated ECG classification across heterogeneous IoT devices, quantifying both performance and communication efficiency. To evaluate feasibility in realistic IoT settings, we deployed the framework across a server, a laptop, and a resource-constrained Raspberry Pi 4, reflecting edge-cloud integration in IoT ecosystems. Experimental results demonstrate that the FL-GAF model achieves a high classification accuracy of 95.18% in a multi-client setup, significantly outperforming a single-client baseline in both accuracy and training time. Despite the added computational complexity of GAF transformations, the framework maintains efficient resource utilization and communication overhead. These findings highlight the potential of lightweight, privacy-preserving AI for IoT-based healthcare monitoring, supporting scalable and secure edge deployments in smart health systems.

Figures (3)

• Figure 1: Proposed Federated Learning Framework
• Figure 2: Example of transforming a 1D ECG signal (of a sample) to a 2D GAF image: (a) 1D vector, and (b) ECG 2D GAF imageelmir2023ecg.
• Figure 3: Architecture Diagram for the proposed CNN model