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

Classification of EMG signals using wavelet neural network.

Abdulhamit Subasi1, Mustafa Yilmaz, Hasan Riza Ozcalik

  • 1Kahramanmaras Sutcu Imam University, Department of Electrical and Electronics Engineering, 46500 Kahramanmaraş, Turkey. asubasi@ksu.edu.tr

Journal of Neuroscience Methods
|April 20, 2006
PubMed
Summary

Wavelet neural networks (WNN) offer a more accurate method for classifying electromyographic (EMG) signals compared to feedforward error backpropagation artificial neural networks (FEBANN). This advancement aids in diagnosing neuromuscular disorders.

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

The Brief Case: Wound infection following a penetrating and cutting instrument injury: a case report.

Journal of clinical microbiology·2026
Same author

The COPD-augmented R₂CHA₂DS₂-VA score: development and internal assessment for predicting one-year outcomes after transcatheter aortic valve implantation.

BMC cardiovascular disorders·2026
Same author

Free-Floating Spherical Left Atrial Thrombus in a Patient With Rheumatic Mitral Stenosis.

CASE (Philadelphia, Pa.)·2026
Same author

Nanocarrier and probe strategies for nuclear targeting.

International journal of pharmaceutics·2026
Same author

In-hospital mortality predictors of heart failure patients 80 years and older: An observational retrospective single-center experience.

Medicine·2026
Same author

Design and synthesis of fluorescent calix[4]arene-1,8-Naphthalimide derivatives: Transparent biofilms, antioxidant and antimicrobial properties, and real-samples applications.

Food chemistry·2026

Area of Science:

  • Biomedical Engineering
  • Computational Neuroscience
  • Signal Processing

Background:

  • Quantitative analysis of electromyographic (EMG) signals is crucial for diagnosing neuromuscular disorders.
  • Computer-aided EMG equipment has advanced, necessitating efficient signal pattern classification methods.
  • Existing time and frequency domain analyses have limitations in EMG signal classification.

Purpose of the Study:

  • To develop and compare artificial neural network (ANN) classifiers for EMG signal patterns.
  • To evaluate the accuracy of feedforward error backpropagation artificial neural networks (FEBANN) and wavelet neural networks (WNN).
  • To assess the utility of these classifiers in supporting differential diagnosis of neuromuscular diseases.

Main Methods:

  • Utilized an autoregressive (AR) model to extract features from EMG signals.

Related Experiment Videos

  • Developed and implemented FEBANN and WNN based classifiers.
  • Analyzed 1200 motor unit potentials (MUPs) from normal subjects and patients with myopathy and neurogenic disease.

    • Main Results:

    • The WNN classifier achieved a success rate of 90.7%.
    • The FEBANN classifier achieved a success rate of 88%.
    • WNN demonstrated superior performance over FEBANN in EMG signal classification accuracy.

    Conclusions:

    • WNN-based classification is a highly accurate and efficient method for EMG signal pattern analysis.
    • The WNN classifier can significantly support expert decision-making in EMG differential diagnosis.
    • This approach holds promise for improving the diagnosis of neuromuscular disorders.