Finding "Good Views" of Electrocardiogram Signals for Inferring Abnormalities in Cardiac Condition

TL;DR

This paper investigates various methods to define positive samples in contrastive learning for ECG signals, finding that patient-invariant representations significantly improve arrhythmia classification performance.

Contribution

It systematically compares different strategies for positive sample selection in contrastive learning for ECGs, highlighting the effectiveness of patient-invariant representations.

Findings

Patient-invariant representations enhance arrhythmia detection.

Contrastive learning strategies vary in effectiveness based on positive sample definition.

The code implementation is publicly available for reproducibility.

Abstract

Electrocardiograms (ECGs) are an established technique to screen for abnormal cardiac signals. Recent work has established that it is possible to detect arrhythmia directly from the ECG signal using deep learning algorithms. While a few prior approaches with contrastive learning have been successful, the best way to define a positive sample remains an open question. In this project, we investigate several ways to define positive samples, and assess which approach yields the best performance in a downstream task of classifying arrhythmia. We explore spatiotemporal invariances, generic augmentations, demographic similarities, cardiac rhythms, and wave attributes of ECG as potential ways to match positive samples. We then evaluate each strategy with downstream task performance, and find that learned representations invariant to patient identity are powerful in arrhythmia detection. We made our code available in: https://github.com/mandiehyewon/goodviews_ecg.git