Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Application of federated learning techniques for arrhythmia classification using 12-lead ECG signals

Daniel Mauricio Jimenez Gutierrez, Hafiz Muuhammad Hassan, Lorella Landi, Andrea Vitaletti, Ioannis Chatzigiannakis

TL;DR

This work uses a Federated Learning (FL) privacy-preserving methodology to train AI models over heterogeneous sets of high-definition ECG from 12-lead sensor arrays collected from six heterogeneous sources, and demonstrates comparable performance to models trained using CL, IID, and non-IID approaches.

Abstract

Artificial Intelligence-based (AI) analysis of large, curated medical datasets is promising for providing early detection, faster diagnosis, and more effective treatment using low-power Electrocardiography (ECG) monitoring devices information. However, accessing sensitive medical data from diverse sources is highly restricted since improper use, unsafe storage, or data leakage could violate a person's privacy. This work uses a Federated Learning (FL) privacy-preserving methodology to train AI models over heterogeneous sets of high-definition ECG from 12-lead sensor arrays collected from six heterogeneous sources. We evaluated the capacity of the resulting models to achieve equivalent performance compared to state-of-the-art models trained in a Centralized Learning (CL) fashion. Moreover, we assessed the performance of our solution over Independent and Identical distributed (IID) and non-IID federated data. Our methodology involves machine learning techniques based on Deep Neural Networks and Long-Short-Term Memory models. It has a robust data preprocessing pipeline with feature engineering, selection, and data balancing techniques. Our AI models demonstrated comparable performance to models trained using CL, IID, and non-IID approaches. They showcased advantages in reduced complexity and faster training time, making them well-suited for cloud-edge architectures.

Application of federated learning techniques for arrhythmia classification using 12-lead ECG signals

TL;DR

This work uses a Federated Learning (FL) privacy-preserving methodology to train AI models over heterogeneous sets of high-definition ECG from 12-lead sensor arrays collected from six heterogeneous sources, and demonstrates comparable performance to models trained using CL, IID, and non-IID approaches.

Abstract

Artificial Intelligence-based (AI) analysis of large, curated medical datasets is promising for providing early detection, faster diagnosis, and more effective treatment using low-power Electrocardiography (ECG) monitoring devices information. However, accessing sensitive medical data from diverse sources is highly restricted since improper use, unsafe storage, or data leakage could violate a person's privacy. This work uses a Federated Learning (FL) privacy-preserving methodology to train AI models over heterogeneous sets of high-definition ECG from 12-lead sensor arrays collected from six heterogeneous sources. We evaluated the capacity of the resulting models to achieve equivalent performance compared to state-of-the-art models trained in a Centralized Learning (CL) fashion. Moreover, we assessed the performance of our solution over Independent and Identical distributed (IID) and non-IID federated data. Our methodology involves machine learning techniques based on Deep Neural Networks and Long-Short-Term Memory models. It has a robust data preprocessing pipeline with feature engineering, selection, and data balancing techniques. Our AI models demonstrated comparable performance to models trained using CL, IID, and non-IID approaches. They showcased advantages in reduced complexity and faster training time, making them well-suited for cloud-edge architectures.
Paper Structure (26 sections, 2 equations, 15 figures, 7 tables)

This paper contains 26 sections, 2 equations, 15 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Previous related work
  3. Arrythmia detection
  4. Federated learning
  5. Methodology
  6. Problem Formulation
  7. Feature Extraction
  8. Feature Normalization
  9. Feature Selection
  10. Dataset Balancing
  11. Multiclass Arrhythmia Classification
  12. Model 1 - (DNN) Deep Neural Networks
  13. Model 2 - (LSTM) Long-Short Term Memory
  14. Distributed Weight Aggregation
  15. Experimental settings and fine-tuning
  16. ...and 11 more sections

Figures (15)

  • Figure 1: High-level overview of the proposed federated learning methodology and software and hardware components.
  • Figure 2: Typical ECG waveform from the Lead II position for one cardiac cycle.
  • Figure 3: DNN architecture
  • Figure 4: LSTM architecture
  • Figure 5: Feature importance from XG-Boost algorithm (only the 50 best)
  • ...and 10 more figures