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

The forward EEG solutions can be computed using artificial neural networks.

M Sun1, R J Sclabassi

  • 1Department of Neurological Surgery and Biomedical Engineering, University of Pittsburgh, PA 15213, USA. mrsun@neuronet.pitt.edu

IEEE Transactions on Bio-Medical Engineering
|August 16, 2000
PubMed
Summary

This study introduces an artificial neural network (ANN) to accurately model electroencephalography (EEG) brain waves. The ANN efficiently computes forward solutions for brain activity localization, improving upon traditional methods.

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

Comparative study on effect of different promoters on expression of cry1Ac in Bacillus thuringiensis chromosome.

Journal of applied microbiology·2007
Same author

A surging new interest on toilet training started during the first months of age in Western countries.

Techniques in coloproctology·2007
Same author

Endoscopic ultrasound-guided interstitial chemotherapy in the pancreas: results in a canine model.

Endoscopy·2007
Same author

Expression of breast cancer resistance protein (BCRP/ABCG2) in human placenta throughout gestation and at term before and after labor.

Canadian journal of physiology and pharmacology·2007
Same author

Herbicide exposure modifies GSTP1 haplotype association to Parkinson onset age: the GenePD Study.

Neurology·2006
Same author

c-Jun N-terminal kinase is activated in non-small-cell lung cancer and promotes neoplastic transformation in human bronchial epithelial cells.

Oncogene·2006

Area of Science:

  • Computational neuroscience
  • Biomedical engineering
  • Signal processing

Background:

  • Electroencephalography (EEG) is crucial for brain research and neurological diagnosis.
  • Large electrode arrays enable detailed EEG recording.
  • Head volume conductor models aid in localizing brain activity.
  • Traditional methods (BEM, FEM) for computing forward EEG solutions are computationally intensive.

Purpose of the Study:

  • To develop a novel computational approach using artificial neural networks (ANNs) for efficient and accurate EEG forward solutions.
  • To map forward solutions between simplified (spherical) and complex (spheroidal) head models.
  • To overcome the computational limitations of traditional BEM and FEM methods.

Main Methods:

Related Experiment Videos

  • An artificial neural network (ANN) was trained to map forward solutions from a computationally efficient spherical head model to a more precise spheroidal head model.
  • The ANN input vector was derived from the spherical model, and the output vector represented the spheroidal model solution.
  • The performance of the ANN approach was evaluated against exact solutions and compared with BEM and FEM methods.

    • Main Results:

    • The ANN approach achieved a mean-square error of approximately 0.3% compared to exact solutions.
    • Online computation was highly efficient, requiring only 168 floating point operations per channel.
    • The ANN model required minimal storage (10.2 K-bytes).

    Conclusions:

    • The proposed ANN method offers a significant improvement over BEM and FEM for EEG forward modeling.
    • Real-time and accurate EEG modeling on personal computers is feasible using this ANN approach.
    • This method enhances the efficiency and accessibility of brain activity localization techniques.