Multivariate Bayesian classification of epilepsy EEG signals

Antonio Quintero-Rincón, Jorge Prendes, Marcelo Pereyra, Hadj Batatia, Marcelo Risk

Research output: Chapter in Book/Report/Conference proceedingConference contribution

10 Citations (Scopus)

Abstract

The classification of epileptic seizure events in EEG signals is an important problem in biomedical engineering. In this paper we propose a Bayesian classification method for multivariate EEG signals. The method is based on a multilevel 2D wavelet decomposition that captures the distribution of energy across the different brain rhythms and regions, coupled with a generalised Gaussian statistical representation and a multivariate Bayesian classification scheme. The proposed approach is demonstrated on a challenging paediatric dataset containing both epileptic events and normal brain function signals, where it outperforms a state-of-the-art method both in terms of classification sensitivity and specificity.
Original languageEnglish
Title of host publication2016 IEEE 12th Image, Video, and Multidimensional Signal Processing Workshop (IVMSP)
PublisherIEEE
ISBN (Electronic)9781509019298
DOIs
Publication statusPublished - 4 Aug 2016

Keywords

  • Bayes methods
  • Gaussian processes
  • electroencephalography
  • medical signal processing
  • paediatrics
  • signal classification
  • wavelet transforms
  • biomedical engineering
  • brain rhythm
  • epilepsy EEG signals
  • epileptic seizure event classification
  • generalised Gaussian statistical representation
  • multilevel 2D wavelet decomposition
  • multivariate Bayesian classification
  • multivariate EEG signal
  • normal brain function signal
  • paediatric dataset
  • Brain modeling
  • Electroencephalography
  • Epilepsy
  • Gaussian distribution
  • Sensitivity
  • Two dimensional displays
  • Bayesian classifiers
  • EEG
  • Generalized Gaussian distribution
  • Multilevel 2D wavelet

Fingerprint Dive into the research topics of 'Multivariate Bayesian classification of epilepsy EEG signals'. Together they form a unique fingerprint.

Cite this