EEG source localization based on a structured sparsity prior and a partially collapsed Gibbs sampler

Facundo Costa, Hadj Batatia, Thomas Oberlin, Jean-Yves Tourneret

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

3 Citations (Scopus)

Abstract

In this paper, we propose a hierarchical Bayesian model approximating the ℓ20 mixed-norm regularization by a multivariate Bernoulli Laplace prior to solve the EEG inverse problem by promoting spatial structured sparsity. The posterior distribution of this model is too complex to derive closed-form expressions of the standard Bayesian estimators. An MCMC method is proposed to sample this posterior and estimate the model parameters from the generated samples. The algorithm is based on a partially collapsed Gibbs sampler and a dual dipole random shift proposal for the non-zero positions. The brain activity and all other model parameters are jointly estimated in a completely unsupervised framework. The results obtained on synthetic data with controlled ground truth show the good performance of the proposed method when compared to the ℓ21 approach in different scenarios, and its capacity to estimate point-like source activity.

Original languageEnglish
Title of host publication2015 IEEE 6th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP)
PublisherIEEE
Pages261-264
Number of pages4
ISBN (Electronic)9781479919635
DOIs
Publication statusPublished - 21 Jan 2016
Event6th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing 2015 - Cancun, Mexico
Duration: 13 Dec 201516 Dec 2015

Conference

Conference6th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing 2015
Abbreviated titleCAMSAP 2015
Country/TerritoryMexico
CityCancun
Period13/12/1516/12/15

Keywords

  • EEG
  • hierarchical Bayesian model
  • inverse problem
  • MCMC
  • source localization
  • structured-sparsity
  • ℓ20-norm regularization

ASJC Scopus subject areas

  • Signal Processing
  • Computational Mathematics

Fingerprint

Dive into the research topics of 'EEG source localization based on a structured sparsity prior and a partially collapsed Gibbs sampler'. Together they form a unique fingerprint.

Cite this