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 Videos

Serial VCG/ECG analysis using neural networks

M Sunemark1, L Edenbrandt, H Holst

  • 1Department of Clinical Physiology, Lund University, Sweden.

Computers and Biomedical Research, an International Journal
|April 30, 1998
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

Challenges in conducting paediatric trials with off-patent drugs.

Contemporary clinical trials communications·2021
Same author

Higher chlorzoxazone clearance in obese children compared with nonobese peers.

British journal of clinical pharmacology·2018
Same author

Cardiac signal estimation based on the arterial and venous pressure signals of a hemodialysis machine.

Physiological measurement·2016
Same author

Identification of patients prone to hypotension during hemodialysis based on the analysis of cardiovascular signals.

Medical engineering & physics·2015
Same author

A modified Lewis ECG lead system for ambulatory monitoring of atrial arrhythmias.

Journal of electrocardiology·2014
Same author

HIV+ and HIV- youth living in group homes in South Africa need more psychosocial support.

Vulnerable children and youth studies·2013

This study explored artificial neural networks for serial ECG analysis. Combining ECG and VCG measurements improved accuracy for detecting myocardial infarction, but VCG loop alignment did not significantly enhance results.

Area of Science:

  • Cardiology
  • Artificial Intelligence in Medicine
  • Signal Processing

Background:

  • Serial electrocardiogram (ECG) analysis is crucial for diagnosing conditions like myocardial infarction by comparing successive recordings.
  • Challenges in serial ECG analysis include false positive changes due to electrode misplacement or patient positioning.
  • Artificial neural networks (ANNs) offer a potential solution for improving the accuracy of serial ECG interpretation.

Purpose of the Study:

  • To investigate a novel approach using ANNs for serial ECG analysis.
  • To evaluate the effectiveness of VCG loop alignment in compensating for interrecording variability.
  • To compare the diagnostic performance of ANNs using combined ECG and vectorcardiogram (VCG) measurements versus ECG or VCG alone.

Main Methods:

Related Experiment Videos

  • A study population of 1000 patients with two ECG recordings was analyzed.
  • A new technique for VCG loop alignment was employed to address positional changes.
  • ANNs were trained using various combinations of ECG and VCG measurements to detect pathological changes indicative of new infarcts.

Main Results:

  • The best ANN performance, combining ECG and VCG measurements, achieved a sensitivity of 69% and a specificity of 90%.
  • Using only ECG or VCG measurements resulted in lower sensitivity (63% and 60%, respectively).
  • Inclusion of VCG loop alignment did not significantly improve the performance of the neural network-based serial analysis.

Conclusions:

  • ANNs show promise for enhancing serial ECG analysis, particularly when integrating both ECG and VCG data.
  • The combination of ECG and VCG measurements provides superior diagnostic accuracy compared to using either modality alone.
  • VCG loop alignment, while addressing interrecording variability, did not yield significant performance improvements in this ANN-based approach.