Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Electrocardiogram Fundamentals01:28

Electrocardiogram Fundamentals

519
Introduction
An electrocardiogram (ECG) is a diagnostic tool for identifying cardiac conditions such as arrhythmias, conduction abnormalities, and myocardial ischemia.
Definition
An electrocardiogram (ECG) visualizes the heart's electrical activity by tracing the electrical movement associated with each heartbeat on a graph or monitor. As the heart beats, an electrical wave passes through it, correlating with the cardiac cycle events.
Parts of an ECG
An ECG utilizes electrodes on the skin...
519
Electrocardiogram01:29

Electrocardiogram

2.2K
An electrocardiogram (ECG or EKG) is a critical diagnostic tool that records the electrical signals produced by the heart during each heartbeat. This recording is achieved through electrodes placed strategically on the arms, legs, and chest. The electrocardiograph amplifies these signals and produces 12 distinct tracings, offering a comprehensive understanding of the heart's electrical activity.
Three major waveforms are present in a typical ECG recording: the P wave, the QRS complex, and...
2.2K

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

ActiTect: a generalizable machine learning pipeline for REM sleep behavior disorder screening through standardized actigraphy.

NPJ digital medicine·2026
Same author

Smartphone Sensor Battery Consumption: A Standardized and Reproducible Test Protocol.

Sensors (Basel, Switzerland)·2026
Same author

Wearable Movement-Tracking for Prodromal Parkinson's Disease Detection: A Cross-Country Validation Study.

Movement disorders : official journal of the Movement Disorder Society·2026
Same author

Fully Anonymized Digital Health Data Acquisition in a Research Partnership Using a Blinded Deidentification Proxy in the HerzFit App: Implementation Study.

JMIR formative research·2026
Same author

A Comparative Assessment of Accuracy in Video-Based Monocular Human Pose Estimation Frameworks.

IEEE transactions on pattern analysis and machine intelligence·2026
Same author

XGBOrdinal: An XGBoost Extension for Ordinal Data.

Studies in health technology and informatics·2025

Related Experiment Video

Updated: Jun 6, 2025

Real-Time Cardiac Mapping with a Noninvasive Imageless Electrocardiographic Imaging System
10:17

Real-Time Cardiac Mapping with a Noninvasive Imageless Electrocardiographic Imaging System

Published on: April 11, 2025

417

Disentangled representational learning for anomaly detection in single-lead electrocardiogram signals using

Maximilian Kapsecker1, Matthias C Möller2, Stephan M Jonas3

  • 1TUM School of Computation, Information and Technology, Technical University of Munich, Boltzmannstraße 3, Garching bei München, 85748, Bavaria, Germany; Institute for Digital Medicine, University Hospital Bonn, Venusberg-Campus 1, Bonn, 53127, North Rhine-Westphalia, Germany.

Computers in Biology and Medicine
|November 24, 2024
PubMed
Summary

This study introduces a novel explainable AI method for analyzing electrocardiogram (ECG) data from wearable devices. The approach uses disentangled representation learning for accurate, unsupervised anomaly detection and personalization, improving medical insights.

Keywords:
ElectrocardiographyExplainable anomaly detectionPersonalizationRepresentational learning

More Related Videos

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

957
Analyzing Long-Term Electrocardiography Recordings to Detect Arrhythmias in Mice
06:07

Analyzing Long-Term Electrocardiography Recordings to Detect Arrhythmias in Mice

Published on: May 23, 2021

3.6K

Related Experiment Videos

Last Updated: Jun 6, 2025

Real-Time Cardiac Mapping with a Noninvasive Imageless Electrocardiographic Imaging System
10:17

Real-Time Cardiac Mapping with a Noninvasive Imageless Electrocardiographic Imaging System

Published on: April 11, 2025

417
Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

957
Analyzing Long-Term Electrocardiography Recordings to Detect Arrhythmias in Mice
06:07

Analyzing Long-Term Electrocardiography Recordings to Detect Arrhythmias in Mice

Published on: May 23, 2021

3.6K

Area of Science:

  • Cardiology
  • Artificial Intelligence
  • Signal Processing

Background:

  • Wearable devices enable unsupervised electrocardiogram (ECG) signal recording, but analyzing high-dimensional ECG data presents statistical and explainability challenges.
  • Existing anomaly detection methods for ECGs often lack transparency and struggle with inter-subject variability.

Purpose of the Study:

  • To investigate the feasibility of medically explainable anomaly detection in ECGs using disentangled representational learning.
  • To personalize ECG analysis and mitigate variations between individuals for improved accuracy.

Main Methods:

  • Five open-source ECG datasets were processed into denoised one-second lead I signal traces.
  • A beta total correlation variational autoencoder was optimized, revealing disentanglement between atrial and ventricular features in a 12-dimensional embedding space.
  • A k-nearest neighbor classifier was used for anomaly detection within the learned embedding space, with model fine-tuning for personalization.

Main Results:

  • The model achieved an F1 score of 0.94 for predicting sinus rhythm versus pathological classes (bundle branch blocks, myocardial infarction, AV block).
  • Anomaly detection accuracy reached 90.94%, comparable to established detectors but with enhanced explainability and unsupervised learning.
  • Personalized model fine-tuning yielded an F1 score of 0.93 for predicting normal, premature atrial contraction, and premature ventricular contraction, with consistent pathology distribution plots.

Conclusions:

  • The disentangled variational autoencoder offers a robust and explainable method for ECG representation and anomaly detection.
  • The approach effectively mitigates inter-subject variations through personalization, enhancing diagnostic potential.
  • This method provides medically interpretable insights from unsupervised ECG analysis, advancing wearable healthcare technology.