Intra-Inter Subject Self-supervised Learning for Multivariate Cardiac Signals

TL;DR

This paper introduces a novel self-supervised learning model tailored for multivariate cardiac signals, effectively capturing intra- and inter-subject differences to improve abnormal heart rhythm detection, especially with limited labeled data.

Contribution

The paper proposes the Intra-inter Subject self-supervised Learning (ISL) model that integrates medical knowledge and contrastive learning for better representation of cardiac signals.

Findings

About 10% improvement over supervised methods with 1% labeled data.

Effective learning of intra- and inter-subject differences.

Strong generalizability and robustness demonstrated.

Abstract

Learning information-rich and generalizable representations effectively from unlabeled multivariate cardiac signals to identify abnormal heart rhythms (cardiac arrhythmias) is valuable in real-world clinical settings but often challenging due to its complex temporal dynamics. Cardiac arrhythmias can vary significantly in temporal patterns even for the same patient ($i.e.$, intra subject difference). Meanwhile, the same type of cardiac arrhythmia can show different temporal patterns among different patients due to different cardiac structures ($i.e.$, inter subject difference). In this paper, we address the challenges by proposing an Intra-inter Subject self-supervised Learning (ISL) model that is customized for multivariate cardiac signals. Our proposed ISL model integrates medical knowledge into self-supervision to effectively learn from intra-inter subject differences. In intra subject self-supervision, ISL model first extracts heartbeat-level features from each subject using a channel-wise attentional CNN-RNN encoder. Then a stationarity test module is employed to capture the temporal dependencies between heartbeats. In inter subject self-supervision, we design a set of data augmentations according to the clinical characteristics of cardiac signals and perform contrastive learning among subjects to learn distinctive representations for various types of patients. Extensive experiments on three real-world datasets were conducted. In a semi-supervised transfer learning scenario, our pre-trained ISL model leads about 10% improvement over supervised training when only 1% labeled data is available, suggesting strong generalizability and robustness of the model.