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

Neural network based classification of single-trial EEG data

N Masic1, G Pfurtscheller

  • 1Department of Medical Informatics, Graz University of Technology, Austria.

Artificial Intelligence in Medicine
|December 1, 1993
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

MRI-related anxiety can induce slow BOLD oscillations coupled with cardiac oscillations.

Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology·2021
Same author

Discussion of "time-frequency techniques in biomedical signal analysis: a tutorial review of similarities and differences".

Methods of information in medicine·2013
Same author

Toward a hybrid brain-computer interface based on imagined movement and visual attention.

Journal of neural engineering·2010
Same author

Discrimination of motor imagery-induced EEG patterns in patients with complete spinal cord injury.

Computational intelligence and neuroscience·2009
Same author

Could the beta rebound in the EEG be suitable to realize a "brain switch"?

Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology·2008
Same author

Short-lived brain state after cued motor imagery in naive subjects.

The European journal of neuroscience·2008

Cascade-Correlation neural networks effectively predict finger movement side using single-trial electroencephalography (EEG) data. This method accurately classifies spatio-temporal EEG patterns, highlighting alpha band dynamics preceding hand movements.

Area of Science:

  • Neuroscience
  • Machine Learning
  • Biomedical Engineering

Background:

  • Predicting motor intentions from brain activity is crucial for brain-computer interfaces.
  • Electroencephalography (EEG) offers a non-invasive window into neural dynamics preceding movement.
  • Classifying single-trial EEG data presents challenges due to inherent noise and variability.

Purpose of the Study:

  • To evaluate the efficacy of different neural network architectures for predicting finger movement side.
  • To determine if spatio-temporal patterns in pre-movement EEG data can be reliably classified.
  • To identify the most suitable neural network for analyzing single-trial EEG for motor intention decoding.

Main Methods:

  • Utilized Standard Back Propagation (BP), Partially Recurrent (PR), and Cascade-Correlation (CC) neural networks.

Related Experiment Videos

  • Employed non-averaged, single-trial, multi-channel EEG data recorded immediately before finger movement execution.
  • Calculated power values from EEG data as input features for classification.

    • Main Results:

    • The Cascade-Correlation (CC) neural network demonstrated superior performance in classifying EEG patterns.
    • CC networks proved effective in identifying underlying dynamics within the alpha band preceding hand movements.
    • The CC network was found to be fast, stable, and adept at recognizing spatio-temporal EEG features.

    Conclusions:

    • Cascade-Correlation neural networks are a suitable choice for classifying single-trial EEG patterns for motor intention prediction.
    • Analysis of pre-movement alpha band activity is key for decoding movement side.
    • This approach holds promise for advancing brain-computer interface technologies through accurate EEG-based prediction.