Feature Selection Strategy for Classification of Single-Trial EEG Elicited by Motor Imagery

Swati Prasad, Zheng-Hua Tan, Ramjee Prasad, Alvaro Rodrigo Cabrera, Ying Gu, Kim Dremstrup

Research output: Contribution to journalConference article in JournalResearchpeer-review

4 Citations (Scopus)

Abstract

Brain-Computer Interface (BCI) provides new means of communication for people with motor disabilities by utilizing electroencephalographic activity. Selection of features from Electroencephalogram (EEG) signals for classification plays a key part in the development of BCI systems. In this paper, we present a feature selection strategy consisting of channel selection by fisher ratio analysis in the frequency domain and time segment selection by visual inspection in time domain. The proposed strategy achieves an absolute improvement of 7.5% in the misclassification rate as compared with the baseline system that uses wavelet coefficients as features and support vector machine (SVM) as classifier
Original languageEnglish
JournalProceedings of the Wireless Personal Multimedia Communications Symposia
Pages (from-to)1-4
Number of pages4
ISSN1347-6890
Publication statusPublished - 2011
EventThe 14th International Symposium on Wireless Personal Multimedia Communications - Brest, France
Duration: 3 Oct 20116 Oct 2011

Conference

ConferenceThe 14th International Symposium on Wireless Personal Multimedia Communications
CountryFrance
CityBrest
Period03/10/201106/10/2011

Fingerprint

Brain computer interface
Electroencephalography
Feature extraction
Support vector machines
Inspection
Communication

Keywords

  • Brain-computer interface, feature selection, EEG classification, discrete wavelet transform, support vector machine.

Cite this

@inproceedings{a98eed8a3da14e67934aade21b3d8366,
title = "Feature Selection Strategy for Classification of Single-Trial EEG Elicited by Motor Imagery",
abstract = "Brain-Computer Interface (BCI) provides new means of communication for people with motor disabilities by utilizing electroencephalographic activity. Selection of features from Electroencephalogram (EEG) signals for classification plays a key part in the development of BCI systems. In this paper, we present a feature selection strategy consisting of channel selection by fisher ratio analysis in the frequency domain and time segment selection by visual inspection in time domain. The proposed strategy achieves an absolute improvement of 7.5{\%} in the misclassification rate as compared with the baseline system that uses wavelet coefficients as features and support vector machine (SVM) as classifier",
keywords = "Brain-computer interface, feature selection, EEG classification, discrete wavelet transform, support vector machine.",
author = "Swati Prasad and Zheng-Hua Tan and Ramjee Prasad and Cabrera, {Alvaro Rodrigo} and Ying Gu and Kim Dremstrup",
year = "2011",
language = "English",
pages = "1--4",
journal = "Proceedings of the Wireless Personal Multimedia Communications Symposia",
issn = "1347-6890",
publisher = "Causal Productions",

}

Feature Selection Strategy for Classification of Single-Trial EEG Elicited by Motor Imagery. / Prasad, Swati; Tan, Zheng-Hua; Prasad, Ramjee; Cabrera, Alvaro Rodrigo; Gu, Ying; Dremstrup, Kim.

In: Proceedings of the Wireless Personal Multimedia Communications Symposia, 2011, p. 1-4.

Research output: Contribution to journalConference article in JournalResearchpeer-review

TY - GEN

T1 - Feature Selection Strategy for Classification of Single-Trial EEG Elicited by Motor Imagery

AU - Prasad, Swati

AU - Tan, Zheng-Hua

AU - Prasad, Ramjee

AU - Cabrera, Alvaro Rodrigo

AU - Gu, Ying

AU - Dremstrup, Kim

PY - 2011

Y1 - 2011

N2 - Brain-Computer Interface (BCI) provides new means of communication for people with motor disabilities by utilizing electroencephalographic activity. Selection of features from Electroencephalogram (EEG) signals for classification plays a key part in the development of BCI systems. In this paper, we present a feature selection strategy consisting of channel selection by fisher ratio analysis in the frequency domain and time segment selection by visual inspection in time domain. The proposed strategy achieves an absolute improvement of 7.5% in the misclassification rate as compared with the baseline system that uses wavelet coefficients as features and support vector machine (SVM) as classifier

AB - Brain-Computer Interface (BCI) provides new means of communication for people with motor disabilities by utilizing electroencephalographic activity. Selection of features from Electroencephalogram (EEG) signals for classification plays a key part in the development of BCI systems. In this paper, we present a feature selection strategy consisting of channel selection by fisher ratio analysis in the frequency domain and time segment selection by visual inspection in time domain. The proposed strategy achieves an absolute improvement of 7.5% in the misclassification rate as compared with the baseline system that uses wavelet coefficients as features and support vector machine (SVM) as classifier

KW - Brain-computer interface, feature selection, EEG classification, discrete wavelet transform, support vector machine.

M3 - Conference article in Journal

SP - 1

EP - 4

JO - Proceedings of the Wireless Personal Multimedia Communications Symposia

JF - Proceedings of the Wireless Personal Multimedia Communications Symposia

SN - 1347-6890

ER -