Jove
Visualize
Contact Us

Related Experiment Videos

Visual evoked potentials discrimination based on adaptive zero-tracking neural network.

A Mghari1, M M Himmi, A Amaloud

  • 1Département de Physique, Université My Ismail faculté des Sciences et Techniques, Boutalamine, Errachidia BP 509, Morocco. a.mghari@hotmail.com

Computers in Biology and Medicine
|June 28, 2005
PubMed
Summary

A novel non-linear classifier accurately distinguishes visual evoked potentials (VEP) using zero-tracking and neural networks. This method shows higher accuracy than traditional latency techniques for VEP analysis.

Related Concept Videos

You might also read

Related Articles

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

Sort by
Same journal

Corrigendum to "Visual modalities-based multimodal fusion for surgical phase recognition" [Comput. Biol. Med. 166 (2023) 107453].

Computers in biology and medicine·2026
Same journal

Hybrid cellular automaton-based model for quorum sensing-controlled biofilm evolution.

Computers in biology and medicine·2026
Same journal

Isobonducellin phytocompound from Artemisia annua L. plant exhibits anti-colorectal cancer activity via AKT1 regulation: an in-silico study.

Computers in biology and medicine·2026
Same journal

A broken power-law model of heart rate variability spectra in sleep.

Computers in biology and medicine·2026
Same journal

Muscle synergy adaptation with fatigue development in constant-power cycling.

Computers in biology and medicine·2026
Same journal

Corrigendum to "Integrative adaptive indexes from noisy routine haematological markers can predict and discriminate health status and biological age" [Comput. Biol. Med. 208 (2026) 111628].

Computers in biology and medicine·2026
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

Area of Science:

  • Neuroscience
  • Biomedical Engineering
  • Machine Learning

Background:

  • Visual evoked potentials (VEP) are crucial for assessing visual pathway function.
  • Current hospital methods, like the latency method, have limitations in accuracy.
  • Developing advanced classifiers for VEP analysis is essential for improved diagnostics.

Purpose of the Study:

  • To propose and evaluate a novel non-linear classifier for discriminating visual evoked potentials (VEP).
  • To combine the zero-tracking method with a multi-layer neural network for enhanced VEP feature extraction and classification.
  • To compare the performance of the proposed classifier against the conventional latency method.

Main Methods:

  • VEP data was processed using an adaptive linear prediction filter within the zero-tracking method to extract feature vectors.

Related Experiment Videos

  • A multi-layer neural network, specifically a back-propagation network, was employed for classification.
  • The classifier was tested on 105 VEPs from 48 healthy individuals and 57 patients.

    • Main Results:

    • The proposed non-linear classifier achieved a total success rate of 89%.
    • The back-propagation network demonstrated robust performance in discriminating VEPs.
    • The classifier proved to be more accurate than the standard latency method used in clinical settings.

    Conclusions:

    • The combined zero-tracking method and multi-layer network offer a highly effective approach for VEP discrimination.
    • This advanced classification technique shows significant potential for improving the accuracy of visual pathway assessments.
    • The proposed method represents a promising alternative to traditional VEP analysis techniques in hospitals.