Bayesian nonstationary autoregressive models for biomedical signal analysis

Michael J Cassidy, William D Penny

Research output: Contribution to journalArticlepeer-review

44 Citations (Scopus)

Abstract

We describe a variational Bayesian algorithm for the estimation of a multivariate autoregressive model with time-varying coefficients that adapt according to a linear dynamical system. The algorithm allows for time and frequency domain characterization of nonstationary multivariate signals and is especially suited to the analysis of event-related data. Results are presented on synthetic data and real electroencephalogram data recorded in event-related desynchronization and photic synchronization scenarios.

Original languageEnglish
Pages (from-to)1142-1152
Number of pages11
JournalIEEE Transactions on Biomedical Engineering
Volume49
Issue number10
DOIs
Publication statusPublished - Oct 2002

Keywords

  • Algorithms
  • Bayes Theorem
  • Computer Simulation
  • Electroencephalography
  • Visual Evoked Potentials
  • Humans
  • Likelihood Functions
  • Linear Models
  • Statistical Models
  • Quality Control
  • Regression Analysis
  • Computer-Assisted Signal Processing
  • Time Factors
  • Comparative Study

Cite this