Blind source separation by ICA for EEG multiple sources localization

Yongjian Chen; Qinyu Zhang; Yohsuke Kinouchi

doi:10.1007/978-3-540-36841-0_696

Blind source separation by ICA for EEG multiple sources localization

Yongjian Chen^*
, Qinyu Zhang
, Yohsuke Kinouchi

^*Corresponding author for this work

Tokushima University
Harbin Institute of Technology Shenzhen

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

4 Scopus citations

Abstract

In this paper we describe that Independent Component Analysis (ICA) method for computing the brain signals of unknown source parameters for the inverse problem. We apply Blind Source Separation (BSS) based on ICA for separating multichannel EEG evoked by multiple dipoles into temporally independent stationary sources. For every independent source, we manage to know electrode potentials evoked by every dipole separately by the projection of independent activation maps back onto the electrode arrays. Then for every set of electrode potentials, we need to perform a source localization procedure, and search only for one dipole, thus dramatically reducing the search complexity. In the paper, it is explored that the possibility of applying ICA for EEG multiple dipoles localization when the data are corrupted by additive noise. Before ICA processing, we apply a method to estimate the dipoles number beforehand and reduce dimensionality that can reduce the ICA complexity and improve the unmixing accuracy.

Original language	English
Title of host publication	IFMBE Proceedings
Editors	Sun I. Kim, Tae Suk Suh
Publisher	Springer Verlag
Pages	2760-2763
Number of pages	4
Edition	1
ISBN (Print)	9783540368397
DOIs	https://doi.org/10.1007/978-3-540-36841-0_696
State	Published - 2007
Externally published	Yes
Event	10th World Congress on Medical Physics and Biomedical Engineering, WC 2006 - Seoul, Korea, Republic of Duration: 27 Aug 2006 → 1 Sep 2006

Publication series

Name	IFMBE Proceedings
Number	1
Volume	14
ISSN (Print)	1680-0737
ISSN (Electronic)	1433-9277

Conference

Conference	10th World Congress on Medical Physics and Biomedical Engineering, WC 2006
Country/Territory	Korea, Republic of
City	Seoul
Period	27/08/06 → 1/09/06

Keywords

Blind source separation
Independent component analysis
Inverse problem
Principal component analysis
Sources number estimation

Access to Document

10.1007/978-3-540-36841-0_696

Cite this

@inproceedings{dffd4aa1e83245899280f0fe1975c3a7,

title = "Blind source separation by ICA for EEG multiple sources localization",

abstract = "In this paper we describe that Independent Component Analysis (ICA) method for computing the brain signals of unknown source parameters for the inverse problem. We apply Blind Source Separation (BSS) based on ICA for separating multichannel EEG evoked by multiple dipoles into temporally independent stationary sources. For every independent source, we manage to know electrode potentials evoked by every dipole separately by the projection of independent activation maps back onto the electrode arrays. Then for every set of electrode potentials, we need to perform a source localization procedure, and search only for one dipole, thus dramatically reducing the search complexity. In the paper, it is explored that the possibility of applying ICA for EEG multiple dipoles localization when the data are corrupted by additive noise. Before ICA processing, we apply a method to estimate the dipoles number beforehand and reduce dimensionality that can reduce the ICA complexity and improve the unmixing accuracy.",

keywords = "Blind source separation, Independent component analysis, Inverse problem, Principal component analysis, Sources number estimation",

author = "Yongjian Chen and Qinyu Zhang and Yohsuke Kinouchi",

note = "Publisher Copyright: {\textcopyright} International Federation for Medical and Biological Engineering 2007.; 10th World Congress on Medical Physics and Biomedical Engineering, WC 2006 ; Conference date: 27-08-2006 Through 01-09-2006",

year = "2007",

doi = "10.1007/978-3-540-36841-0\_696",

language = "英语",

isbn = "9783540368397",

series = "IFMBE Proceedings",

publisher = "Springer Verlag",

number = "1",

pages = "2760--2763",

editor = "Kim, \{Sun I.\} and Suh, \{Tae Suk\}",

booktitle = "IFMBE Proceedings",

address = "德国",

edition = "1",

}

Chen, Y, Zhang, Q & Kinouchi, Y 2007, Blind source separation by ICA for EEG multiple sources localization. in SI Kim & TS Suh (eds), IFMBE Proceedings. 1 edn, IFMBE Proceedings, no. 1, vol. 14, Springer Verlag, pp. 2760-2763, 10th World Congress on Medical Physics and Biomedical Engineering, WC 2006, Seoul, Korea, Republic of, 27/08/06. https://doi.org/10.1007/978-3-540-36841-0_696

TY - GEN

T1 - Blind source separation by ICA for EEG multiple sources localization

AU - Chen, Yongjian

AU - Zhang, Qinyu

AU - Kinouchi, Yohsuke

N1 - Publisher Copyright: © International Federation for Medical and Biological Engineering 2007.

PY - 2007

Y1 - 2007

N2 - In this paper we describe that Independent Component Analysis (ICA) method for computing the brain signals of unknown source parameters for the inverse problem. We apply Blind Source Separation (BSS) based on ICA for separating multichannel EEG evoked by multiple dipoles into temporally independent stationary sources. For every independent source, we manage to know electrode potentials evoked by every dipole separately by the projection of independent activation maps back onto the electrode arrays. Then for every set of electrode potentials, we need to perform a source localization procedure, and search only for one dipole, thus dramatically reducing the search complexity. In the paper, it is explored that the possibility of applying ICA for EEG multiple dipoles localization when the data are corrupted by additive noise. Before ICA processing, we apply a method to estimate the dipoles number beforehand and reduce dimensionality that can reduce the ICA complexity and improve the unmixing accuracy.

AB - In this paper we describe that Independent Component Analysis (ICA) method for computing the brain signals of unknown source parameters for the inverse problem. We apply Blind Source Separation (BSS) based on ICA for separating multichannel EEG evoked by multiple dipoles into temporally independent stationary sources. For every independent source, we manage to know electrode potentials evoked by every dipole separately by the projection of independent activation maps back onto the electrode arrays. Then for every set of electrode potentials, we need to perform a source localization procedure, and search only for one dipole, thus dramatically reducing the search complexity. In the paper, it is explored that the possibility of applying ICA for EEG multiple dipoles localization when the data are corrupted by additive noise. Before ICA processing, we apply a method to estimate the dipoles number beforehand and reduce dimensionality that can reduce the ICA complexity and improve the unmixing accuracy.

KW - Blind source separation

KW - Independent component analysis

KW - Inverse problem

KW - Principal component analysis

KW - Sources number estimation

UR - https://www.scopus.com/pages/publications/84887708932

U2 - 10.1007/978-3-540-36841-0_696

DO - 10.1007/978-3-540-36841-0_696

M3 - 会议稿件

AN - SCOPUS:84887708932

SN - 9783540368397

T3 - IFMBE Proceedings

SP - 2760

EP - 2763

BT - IFMBE Proceedings

A2 - Kim, Sun I.

A2 - Suh, Tae Suk

PB - Springer Verlag

T2 - 10th World Congress on Medical Physics and Biomedical Engineering, WC 2006

Y2 - 27 August 2006 through 1 September 2006

ER -

Blind source separation by ICA for EEG multiple sources localization

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this