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

Late potential recognition by artificial neural networks

Q Xue1, B R Reddy

  • 1Marquette Electronics Inc., Milwaukee, WI 53223 USA. xue@diageng.mei.com

IEEE Transactions on Bio-Medical Engineering
|February 1, 1997
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

Draft whole-genome and mitochondrial genome assemblies of <i>Steinernema tarimense</i> and <i>Heterorhabditis</i> sp. XJ-55.

Journal of helminthology·2026
Same author

[A case of invasive fungal rhinosinusitis complicated with intracranial Aspergillosis presenting with blindness as the main complaint].

Zhonghua er bi yan hou tou jing wai ke za zhi = Chinese journal of otorhinolaryngology head and neck surgery·2026
Same author

The Association between Dietary Protein Diversity and Protein Patterns with Frailty in Older Chinese Adults: A Population-Based Cohort Study.

The journal of nutrition, health & aging·2023
Same author

[Classification, diagnosis and treatment status of pulmonary hypertension from 2012 to 2019: a single center study in Yunnan province].

Zhonghua xin xue guan bing za zhi·2023
Same author

Association between Elevated Magnesium Intake and Reduced Risk of Recurrent Falls and Frailty in Osteoarthritis: Data from the Osteoarthritis Initiative.

The journal of nutrition, health & aging·2023
Same author

The efficacy of hydrogel containing zinc oxide-loaded and minocycline serum albumin nanopartical in the treatment of peri-implantitis.

Medicina oral, patologia oral y cirugia bucal·2023

Artificial neural networks effectively identify ventricular tachycardia risk using electrocardiogram signals. This advanced method improves upon traditional analyses by incorporating waveform morphology for more accurate patient stratification.

Area of Science:

  • Cardiology
  • Biomedical Engineering
  • Artificial Intelligence

Background:

  • Ventricular late potentials (LP's) on signal-averaged electrocardiograms (SAECG's) help identify patients at risk for ventricular tachycardia (VT).
  • Accurate identification of patients with inducible VT during electrophysiology testing is crucial for risk stratification and management.

Purpose of the Study:

  • To develop and evaluate an artificial neural network (ANN) model for distinguishing patients with positive electrophysiology (PEP) tests from those with negative electrophysiology (NEP) tests using LP's.
  • To enhance the diagnostic accuracy of SAECG analysis by incorporating vector magnitude waveform morphology alongside traditional time-domain features.

Main Methods:

  • A combined feature set including total QRS duration (TQRSD), high-frequency low-amplitude signal duration (HFLAD), root-mean-square voltage (RMSV), and vector magnitude waveform morphology was used.

Related Experiment Videos

  • A self-organizing and supervised ANN model was developed and compared against a Bayesian classification model.
  • A fuzzy training set was created by introducing random shifts to the QRS offset point to improve model robustness against detection errors.

    • Main Results:

    • The ANN model utilizing the combined feature set demonstrated superior pattern recognition performance compared to the Bayesian model using only time-domain features.
    • Nonlinear transformation within the ANN's hidden layer effectively increased the Euclidean distance between PEP and NEP patterns, enhancing class separability.
    • The developed ANN model showed improved accuracy in identifying patients with inducible VT.

    Conclusions:

    • Artificial neural networks, particularly when incorporating morphological features of vector magnitude waveforms, offer a powerful tool for analyzing ventricular late potentials.
    • This approach enhances the ability to differentiate between patients with and without inducible ventricular tachycardia, potentially improving clinical risk assessment.
    • The study highlights the benefit of advanced signal processing and machine learning techniques in interpreting complex electrocardiogram data for improved cardiac patient management.