Investigation of methods for extracting features related to motor imagery and resting states in EEG-based BCI system

I. Putu Susila; Shin'Ichiro Kanoh; Ko Ichiro Miyamoto; Tatsuo Yoshinobu

doi:10.1541/ieejeiss.129.1828

Investigation of methods for extracting features related to motor imagery and resting states in EEG-based BCI system

I. Putu Susila, Shin'Ichiro Kanoh, Ko Ichiro Miyamoto, Tatsuo Yoshinobu

Research output: Contribution to journal › Article › peer-review

1 Citation (Scopus)

Abstract

Methods for extracting features of motor imagery from 1-channel bipolar EEG were evaluated. The EEG power spectrums which were used as feature vectors were calculated with filter bank, FFT and AR model, and were then classified by linear discriminant analysis (LDA) to discriminate motor imagery and resting states. It was shown that the extraction method using AR model gave the best result with the average true positive rate of 83% (a -7%). Furthermore, when principal component analysis (PCA) was applied to the feature vectors, the dimension of the feature vectors could be reduced without decreasing accuracy of discrimination.

Original language	English
Pages (from-to)	1828-1833
Number of pages	6
Journal	IEEJ Transactions on Electronics, Information and Systems
Volume	129
Issue number	10
DOIs	https://doi.org/10.1541/ieejeiss.129.1828
Publication status	Published - 2009

Keywords

Brain-computer interface (BCI)
Feature extraction
Linear discriminant analysis (LDA), principal component analysis (PCA)
Motor imagery

ASJC Scopus subject areas

Electrical and Electronic Engineering

Access to Document

10.1541/ieejeiss.129.1828

Cite this

@article{8d50543e067e42c9ba3811b574a7c121,

title = "Investigation of methods for extracting features related to motor imagery and resting states in EEG-based BCI system",

abstract = "Methods for extracting features of motor imagery from 1-channel bipolar EEG were evaluated. The EEG power spectrums which were used as feature vectors were calculated with filter bank, FFT and AR model, and were then classified by linear discriminant analysis (LDA) to discriminate motor imagery and resting states. It was shown that the extraction method using AR model gave the best result with the average true positive rate of 83% (a -7%). Furthermore, when principal component analysis (PCA) was applied to the feature vectors, the dimension of the feature vectors could be reduced without decreasing accuracy of discrimination.",

keywords = "Brain-computer interface (BCI), Feature extraction, Linear discriminant analysis (LDA), principal component analysis (PCA), Motor imagery",

author = "Susila, {I. Putu} and Shin'Ichiro Kanoh and Miyamoto, {Ko Ichiro} and Tatsuo Yoshinobu",

year = "2009",

doi = "10.1541/ieejeiss.129.1828",

language = "English",

volume = "129",

pages = "1828--1833",

journal = "IEEJ Transactions on Electronics, Information and Systems",

issn = "0385-4221",

publisher = "The Institute of Electrical Engineers of Japan",

number = "10",

}

TY - JOUR

T1 - Investigation of methods for extracting features related to motor imagery and resting states in EEG-based BCI system

AU - Susila, I. Putu

AU - Kanoh, Shin'Ichiro

AU - Miyamoto, Ko Ichiro

AU - Yoshinobu, Tatsuo

PY - 2009

Y1 - 2009

N2 - Methods for extracting features of motor imagery from 1-channel bipolar EEG were evaluated. The EEG power spectrums which were used as feature vectors were calculated with filter bank, FFT and AR model, and were then classified by linear discriminant analysis (LDA) to discriminate motor imagery and resting states. It was shown that the extraction method using AR model gave the best result with the average true positive rate of 83% (a -7%). Furthermore, when principal component analysis (PCA) was applied to the feature vectors, the dimension of the feature vectors could be reduced without decreasing accuracy of discrimination.

AB - Methods for extracting features of motor imagery from 1-channel bipolar EEG were evaluated. The EEG power spectrums which were used as feature vectors were calculated with filter bank, FFT and AR model, and were then classified by linear discriminant analysis (LDA) to discriminate motor imagery and resting states. It was shown that the extraction method using AR model gave the best result with the average true positive rate of 83% (a -7%). Furthermore, when principal component analysis (PCA) was applied to the feature vectors, the dimension of the feature vectors could be reduced without decreasing accuracy of discrimination.

KW - Brain-computer interface (BCI)

KW - Feature extraction

KW - Linear discriminant analysis (LDA), principal component analysis (PCA)

KW - Motor imagery

UR - http://www.scopus.com/inward/record.url?scp=70450219417&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70450219417&partnerID=8YFLogxK

U2 - 10.1541/ieejeiss.129.1828

DO - 10.1541/ieejeiss.129.1828

M3 - Article

AN - SCOPUS:70450219417

SN - 0385-4221

VL - 129

SP - 1828

EP - 1833

JO - IEEJ Transactions on Electronics, Information and Systems

JF - IEEJ Transactions on Electronics, Information and Systems

IS - 10

ER -

Investigation of methods for extracting features related to motor imagery and resting states in EEG-based BCI system

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this