A BISPECTRUM APPROACH TO FEATURE EXTRACTION FOR A MOTOR IMAGERY BASED BRAIN-COMPUTER INTERFACING SYSTEM

Shahjahan Shahid, RK Sinha, G Prasad

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

9 Citations (Scopus)

Abstract

Existing feature extraction techniques for BCI systems are developed based on traditional signal processing techniques assuming that the signal is Gaussian and has linear characteristics.But the motor imagery (MI) related EEG is highly non-Gaussian, non-stationary and non-linear. This paper proposes an advanced, robust but simple feature extraction procedure for MI based BCI system. This novel approach uses higher order statistics technique, the bispectrum, and extracts the non-linear features from EEG. Along with a linear classifier (LDA), the proposed technique has been applied to an MI based BCI system. The performance (classification accuracy, mutual information and Cohens kappa) of the system is evaluated and compared with the power spectrum based BCI. It is observed that the proposed technique extracts more pragmatic information resulting in better and consistent cross-session detection accuracy and Cohens kappa. It is concluded that the bispectrum based feature extraction is a promising technique for detecting different brain states.
LanguageEnglish
Title of host publicationUnknown Host Publication
Pages1831-1835
Number of pages5
Publication statusPublished - 2010
Event18th European Signal Processing Conf. (EUSIPCO-2010), Aalborg, Denmark -
Duration: 1 Jan 2010 → …

Conference

Conference18th European Signal Processing Conf. (EUSIPCO-2010), Aalborg, Denmark
Period1/01/10 → …

Fingerprint

Feature extraction
Brain
Computer systems
Electroencephalography
Higher order statistics
Power spectrum
Signal processing
Classifiers

Cite this

Shahid, Shahjahan ; Sinha, RK ; Prasad, G. / A BISPECTRUM APPROACH TO FEATURE EXTRACTION FOR A MOTOR IMAGERY BASED BRAIN-COMPUTER INTERFACING SYSTEM. Unknown Host Publication. 2010. pp. 1831-1835
@inproceedings{7404ec82da2a48a4a84e0a22c88d9156,
title = "A BISPECTRUM APPROACH TO FEATURE EXTRACTION FOR A MOTOR IMAGERY BASED BRAIN-COMPUTER INTERFACING SYSTEM",
abstract = "Existing feature extraction techniques for BCI systems are developed based on traditional signal processing techniques assuming that the signal is Gaussian and has linear characteristics.But the motor imagery (MI) related EEG is highly non-Gaussian, non-stationary and non-linear. This paper proposes an advanced, robust but simple feature extraction procedure for MI based BCI system. This novel approach uses higher order statistics technique, the bispectrum, and extracts the non-linear features from EEG. Along with a linear classifier (LDA), the proposed technique has been applied to an MI based BCI system. The performance (classification accuracy, mutual information and Cohens kappa) of the system is evaluated and compared with the power spectrum based BCI. It is observed that the proposed technique extracts more pragmatic information resulting in better and consistent cross-session detection accuracy and Cohens kappa. It is concluded that the bispectrum based feature extraction is a promising technique for detecting different brain states.",
author = "Shahjahan Shahid and RK Sinha and G Prasad",
year = "2010",
language = "English",
pages = "1831--1835",
booktitle = "Unknown Host Publication",

}

Shahid, S, Sinha, RK & Prasad, G 2010, A BISPECTRUM APPROACH TO FEATURE EXTRACTION FOR A MOTOR IMAGERY BASED BRAIN-COMPUTER INTERFACING SYSTEM. in Unknown Host Publication. pp. 1831-1835, 18th European Signal Processing Conf. (EUSIPCO-2010), Aalborg, Denmark, 1/01/10.

A BISPECTRUM APPROACH TO FEATURE EXTRACTION FOR A MOTOR IMAGERY BASED BRAIN-COMPUTER INTERFACING SYSTEM. / Shahid, Shahjahan; Sinha, RK; Prasad, G.

Unknown Host Publication. 2010. p. 1831-1835.

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

TY - GEN

T1 - A BISPECTRUM APPROACH TO FEATURE EXTRACTION FOR A MOTOR IMAGERY BASED BRAIN-COMPUTER INTERFACING SYSTEM

AU - Shahid, Shahjahan

AU - Sinha, RK

AU - Prasad, G

PY - 2010

Y1 - 2010

N2 - Existing feature extraction techniques for BCI systems are developed based on traditional signal processing techniques assuming that the signal is Gaussian and has linear characteristics.But the motor imagery (MI) related EEG is highly non-Gaussian, non-stationary and non-linear. This paper proposes an advanced, robust but simple feature extraction procedure for MI based BCI system. This novel approach uses higher order statistics technique, the bispectrum, and extracts the non-linear features from EEG. Along with a linear classifier (LDA), the proposed technique has been applied to an MI based BCI system. The performance (classification accuracy, mutual information and Cohens kappa) of the system is evaluated and compared with the power spectrum based BCI. It is observed that the proposed technique extracts more pragmatic information resulting in better and consistent cross-session detection accuracy and Cohens kappa. It is concluded that the bispectrum based feature extraction is a promising technique for detecting different brain states.

AB - Existing feature extraction techniques for BCI systems are developed based on traditional signal processing techniques assuming that the signal is Gaussian and has linear characteristics.But the motor imagery (MI) related EEG is highly non-Gaussian, non-stationary and non-linear. This paper proposes an advanced, robust but simple feature extraction procedure for MI based BCI system. This novel approach uses higher order statistics technique, the bispectrum, and extracts the non-linear features from EEG. Along with a linear classifier (LDA), the proposed technique has been applied to an MI based BCI system. The performance (classification accuracy, mutual information and Cohens kappa) of the system is evaluated and compared with the power spectrum based BCI. It is observed that the proposed technique extracts more pragmatic information resulting in better and consistent cross-session detection accuracy and Cohens kappa. It is concluded that the bispectrum based feature extraction is a promising technique for detecting different brain states.

M3 - Conference contribution

SP - 1831

EP - 1835

BT - Unknown Host Publication

ER -