Guiding Masked Representation Learning to Capture Spatio-Temporal Relationship of Electrocardiogram

TL;DR

This paper introduces ST-MEM, a self-supervised learning method that captures the spatio-temporal relationships in ECG data to improve arrhythmia classification, addressing the challenge of limited labeled data.

Contribution

The paper proposes ST-MEM, a novel SSL approach that models spatio-temporal features in ECG signals, outperforming existing methods and demonstrating adaptability to different lead configurations.

Findings

ST-MEM outperforms baseline SSL methods in arrhythmia classification.

ST-MEM effectively captures spatio-temporal relationships in ECG data.

The method is adaptable to various ECG lead combinations.

Abstract

Electrocardiograms (ECG) are widely employed as a diagnostic tool for monitoring electrical signals originating from a heart. Recent machine learning research efforts have focused on the application of screening various diseases using ECG signals. However, adapting to the application of screening disease is challenging in that labeled ECG data are limited. Achieving general representation through self-supervised learning (SSL) is a well-known approach to overcome the scarcity of labeled data; however, a naive application of SSL to ECG data, without considering the spatial-temporal relationships inherent in ECG signals, may yield suboptimal results. In this paper, we introduce ST-MEM (Spatio-Temporal Masked Electrocardiogram Modeling), designed to learn spatio-temporal features by reconstructing masked 12-lead ECG data. ST-MEM outperforms other SSL baseline methods in various experimental settings for arrhythmia classification tasks. Moreover, we demonstrate that ST-MEM is adaptable to various lead combinations. Through quantitative and qualitative analysis, we show a spatio-temporal relationship within ECG data. Our code is available at https://github.com/bakqui/ST-MEM.