English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

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

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