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

Bayesian nonstationary autoregressive models for biomedical signal analysis.

Michael J Cassidy1, William D Penny

  • 1Sobell Department of Neurophysiology, Institute of Neurology, University College London, UK. m.cassidy@ion.ucl.ac.uk

IEEE Transactions on Bio-Medical Engineering
|October 11, 2002
PubMed
Summary

We developed a new algorithm to analyze complex, nonstationary signals using time-varying models. This method is ideal for understanding event-related brain activity from electroencephalogram data.

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

Reproducibility in computational sleep research: a call for action.

Sleep·2023
Same author

The Neural Representation of Prospective Choice during Spatial Planning and Decisions.

PLoS biology·2017
Same author

Modulation of frontal effective connectivity during speech.

NeuroImage·2016
Same author

The right hemisphere supports but does not replace left hemisphere auditory function in patients with persisting aphasia.

Brain : a journal of neurology·2013
Same author

Efficient posterior probability mapping using Savage-Dickey ratios.

PloS one·2013
Same author

The known unknowns: neural representation of second-order uncertainty, and ambiguity.

The Journal of neuroscience : the official journal of the Society for Neuroscience·2011

Area of Science:

  • Neuroscience
  • Signal Processing
  • Statistical Modeling

Background:

  • Analyzing nonstationary multivariate signals is challenging.
  • Existing models may not capture dynamic changes in complex data.
  • Event-related data requires methods that can adapt to changing signal characteristics.

Purpose of the Study:

  • To introduce a variational Bayesian algorithm for estimating multivariate autoregressive models.
  • To enable time and frequency domain analysis of nonstationary signals.
  • To apply the algorithm to analyze event-related electroencephalogram (EEG) data.

Main Methods:

  • Developed a variational Bayesian algorithm.
  • Incorporated time-varying coefficients adapting via a linear dynamical system.

Related Experiment Videos

  • Applied the algorithm to synthetic and real EEG data.

    • Main Results:

    • The algorithm effectively estimates multivariate autoregressive models with adaptive coefficients.
    • Demonstrated capability for time and frequency domain characterization of nonstationary signals.
    • Successfully analyzed EEG data from event-related desynchronization and photic synchronization.

    Conclusions:

    • The proposed algorithm provides a robust framework for analyzing nonstationary multivariate signals.
    • It is particularly well-suited for characterizing dynamic changes in event-related data.
    • Validated through application to synthetic and real-world EEG recordings.