Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Supervised mutual-information based feature selection for motor unit action potential classification

N Sheikholeslami¹, D Stashuk

¹Department of Electrical Engineering, McGill University, Montreal, Quebec, Canada.

Medical & Biological Engineering & Computing

|April 16, 1998

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

Age-related neuromuscular changes affecting human vastus lateralis.

The Journal of physiology·2015

Same author

EMG signal decomposition: how can it be accomplished and used?

Journal of electromyography and kinesiology : official journal of the International Society of Electrophysiological Kinesiology·2001

Same author

Fatigue effects on motor unit activity during submaximal contractions.

Archives of physical medicine and rehabilitation·2000

Same author

The relationship of motor unit size, firing rate and force.

Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology·1999

Same author

Firing rate analysis using decompostion-enhanced spike triggered averaging in the quadriceps femoris.

Muscle & nerve·1998

Same author

Robust supervised classification of motor unit action potentials.

Medical & biological engineering & computing·1998

Same journal

Deep multi-modal features based spatio-temporal video regression for non-invasive hemoglobin estimation.

Medical & biological engineering & computing·2026

Same journal

Reduced mechanical strength correlates with decreased elastin content in aortic intima-media tissue: association with dissection in human ascending aortas.

Medical & biological engineering & computing·2026

Same journal

How plaque morphology and stenosis severity govern stent-artery interaction and deployment outcomes: a computational study.

Medical & biological engineering & computing·2026

Same journal

Investigating a relation between amyloid beta plaque burden and accumulated neurotoxicity caused by amyloid beta oligomers.

Medical & biological engineering & computing·2026

Same journal

A robot-assisted eye positioning method with high precision and repeatability for ocular particle therapy: mechanical and geometric assessment.

Medical & biological engineering & computing·2026

Same journal

Enhanced puncture event detection for teleoperated needle insertion robotic system.

Medical & biological engineering & computing·2026

See all related articles

A novel feature selection method using mutual information efficiently reduces data complexity for electromyography (EMG) signals. This approach achieves high accuracy with fewer features, lowering computational costs for motor unit action potential (MUAP) analysis.

Area of Science:

Biomedical Engineering
Signal Processing
Machine Learning

Background:

Electromyography (EMG) signals contain crucial information about neuromuscular activity.
Motor Unit Action Potential (MUAP) analysis is vital for diagnosing neurological disorders.
Efficient feature selection is necessary to reduce computational load in EMG analysis.

Purpose of the Study:

To introduce a supervised mutual information-based feature selection method for EMG signals.
To evaluate the performance of selected feature subsets against full feature sets and other dimensionality reduction techniques.
To assess the computational efficiency and classification accuracy of the proposed method.

Main Methods:

Utilized real MUAP data from 10 EMG signals.

Related Experiment Videos

Compared feature subsets selected by first-order and second-order mutual information with Linear Discriminant Analysis (LDA) and Principal Component Analysis (PCA).

Evaluated performance using Minimum Euclidean Distance (MED) and a robust classifier.

Main Results:

First-order mutual information selected features (20) achieved error rates comparable to using all 32 samples or LDA/PCA.
Second-order mutual information selected features (15) also yielded similar performance.
First-order feature selection demonstrated significantly lower computational cost than LDA, PCA, and second-order selection.
The robust classifier showed similar average error rates with 20 features as the full set, but with higher assignment rates.

Conclusions:

First-order features provide an efficient, lower-dimensional representation of EMG data.
The proposed method achieves high classification accuracy with reduced computational requirements.
Mutual information-based feature selection offers a computationally advantageous alternative for MUAP analysis.