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 Experiment Video

Updated: May 31, 2025

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

47.8K

Lead-grouped multi-stage learning for myocardial infarction localization.

Lin Guo1, Qianyun Zhan1, Jichao Yang2

  • 1Big Data Institute, Central South University, Changsha, 410083, China.

Methods (San Diego, Calif.)
|January 23, 2025
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Non-invasive Multimodal Cardiovascular Disease Detection Method Based on Comprehensive View Analysis.

Journal of imaging informatics in medicine·2026
Same author

Iron drives protease-independent cleavage of gasdermin D in allergic airway diseases.

Cell·2026
Same author

Small interfering RNA-mediated silencing of mutant NPM1 suppresses acute myeloid leukemia via reversing KAT7 and p300-mediated histone acetylation.

Leukemia·2026
Same author

Luteolin inhibits IL-33/ST2L-induced M2 macrophage polarization in endometriosis.

Molecular immunology·2026
Same author

Mitochondria-Targeted Metal Polyphenol Networks Inhibit Crystalline Nephropathy by Modulating SerpinE1 and Remodeling the Pathological Mineralization Microenvironment.

ACS nano·2026
Same author

Combined targeting of VISTA and sorafenib activates T cell-mediated anti-tumor immunity via the NF-κB/TNF axis in hepatocellular carcinoma.

BMC medicine·2026

This study introduces a novel knowledge-driven method for localizing myocardial infarction (MI) using electrocardiograms (ECGs). The approach enhances diagnostic accuracy by integrating clinical knowledge into ECG analysis, improving patient outcomes.

Area of Science:

  • Cardiology
  • Medical Imaging
  • Artificial Intelligence

Background:

  • Electrocardiograms (ECGs) are vital for diagnosing myocardial infarction (MI), but current analysis methods often neglect crucial clinical knowledge and spatial relationships within multi-lead ECGs.
  • Existing models struggle to comprehensively understand and accurately localize MI due to limitations in feature extraction and spatial relationship utilization.

Purpose of the Study:

  • To develop a knowledge-driven method for improving the accuracy and localization of myocardial infarction (MI) using 12-lead electrocardiograms (ECGs).
  • To address the limitations of current ECG analysis models by incorporating clinical diagnostic knowledge and enhancing spatial feature learning.

Main Methods:

  • A knowledge-driven overlapping lead grouping strategy based on clinical relevance for MI localization.
Keywords:
ECGMI localizationMedical knowledgeMulti-stage learning network

More Related Videos

3D Whole-heart Myocardial Tissue Analysis
06:53

3D Whole-heart Myocardial Tissue Analysis

Published on: April 12, 2017

8.7K

Related Experiment Videos

Last Updated: May 31, 2025

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
10:25

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

47.8K
3D Whole-heart Myocardial Tissue Analysis
06:53

3D Whole-heart Myocardial Tissue Analysis

Published on: April 12, 2017

8.7K
  • A multi-stage deep learning network incorporating convolutional layers, SE-enhanced multi-scale residual blocks, and positional Transformer blocks.
  • A branch-level weighted feature integration mechanism for effective fusion of multi-lead ECG data.

    • Main Results:

    • The proposed method achieved over 80% prediction accuracy in myocardial infarction (MI) localization tasks on the PTB-XL dataset.
    • Demonstrated significant improvements in various metrics compared to existing state-of-the-art methods for ECG-based MI diagnosis.
    • Validated the effectiveness of the knowledge-driven approach and the multi-stage learning network in analyzing complex ECG patterns.

    Conclusions:

    • The developed knowledge-driven overlapping lead grouping and multi-stage learning network significantly enhance the accuracy of myocardial infarction (MI) localization from ECGs.
    • This approach offers a promising advancement in automated ECG analysis, potentially leading to more timely and accurate diagnoses of cardiovascular diseases.
    • The findings highlight the importance of integrating medical knowledge into AI models for improved clinical decision-making in cardiology.