A transform-based feature extraction approach for motor imagery tasks classification

In this paper, we present a new motor imagery classification method in the context of electroencephalography (EEG)-based brain-computer interface (BCI). This method uses a signal-dependent orthogonal transform, referred to as linear prediction singular value decomposition (LP-SVD), for feature extra...

Full description

Bibliographic Details
Main Authors:	Baali, Hamza, Khorshidtalab, Aida, Mesbah, Mustafa, Salami, Momoh Jimoh Eyiomika
Format:	Article
Language:	English
Published:	IEEE 2015
Subjects:	QA75 Electronic computers. Computer science QA76 Computer software
Online Access:	http://irep.iium.edu.my/46812/ http://irep.iium.edu.my/46812/ http://irep.iium.edu.my/46812/ http://irep.iium.edu.my/46812/1/07299634_%281%29.pdf

id	iium-46812
recordtype	eprints
spelling	iium-468122018-06-25T04:27:28Z http://irep.iium.edu.my/46812/ A transform-based feature extraction approach for motor imagery tasks classification Baali, Hamza Khorshidtalab, Aida Mesbah, Mustafa Salami, Momoh Jimoh Eyiomika QA75 Electronic computers. Computer science QA76 Computer software In this paper, we present a new motor imagery classification method in the context of electroencephalography (EEG)-based brain-computer interface (BCI). This method uses a signal-dependent orthogonal transform, referred to as linear prediction singular value decomposition (LP-SVD), for feature extraction. The transform defines the mapping as the left singular vectors of the LP coefficient filter impulse response matrix. Using a logistic tree-based model classifier; the extracted features are classified into one of four motor imagery movements. The proposed approach was first benchmarked against two related state-of-the-art feature extraction approaches, namely, discrete cosine transform (DCT) and adaptive autoregressive (AAR)-based methods. By achieving an accuracy of 67.35%, the LP-SVD approach outperformed the other approaches by large margins (25% compared with DCT and 6 % compared with AAR-based methods). To further improve the discriminatory capability of the extracted features and reduce the computational complexity, we enlarged the extracted feature subset by incorporating two extra features, namely, Q- and the Hotelling's $T^{2}$ statistics of the transformed EEG and introduced a new EEG channel selection method. The performance of the EEG classification based on the expanded feature set and channel selection method was compared with that of a number of the state-of-the-art classification methods previously reported with the BCI IIIa competition data set. Our method came second with an average accuracy of 81.38%. IEEE 2015-10-26 Article PeerReviewed application/pdf en http://irep.iium.edu.my/46812/1/07299634_%281%29.pdf Baali, Hamza and Khorshidtalab, Aida and Mesbah, Mustafa and Salami, Momoh Jimoh Eyiomika (2015) A transform-based feature extraction approach for motor imagery tasks classification. IEEE Journal of Translational Engineering in Health and Medicine, 3. pp. 2100108-1. ISSN 2168-2372 http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=7299634 10.1109/JTEHM.2015.2485261
repository_type	Digital Repository
institution_category	Local University
institution	International Islamic University Malaysia
building	IIUM Repository
collection	Online Access
language	English
topic	QA75 Electronic computers. Computer science QA76 Computer software
spellingShingle	QA75 Electronic computers. Computer science QA76 Computer software Baali, Hamza Khorshidtalab, Aida Mesbah, Mustafa Salami, Momoh Jimoh Eyiomika A transform-based feature extraction approach for motor imagery tasks classification
description	In this paper, we present a new motor imagery classification method in the context of electroencephalography (EEG)-based brain-computer interface (BCI). This method uses a signal-dependent orthogonal transform, referred to as linear prediction singular value decomposition (LP-SVD), for feature extraction. The transform defines the mapping as the left singular vectors of the LP coefficient filter impulse response matrix. Using a logistic tree-based model classifier; the extracted features are classified into one of four motor imagery movements. The proposed approach was first benchmarked against two related state-of-the-art feature extraction approaches, namely, discrete cosine transform (DCT) and adaptive autoregressive (AAR)-based methods. By achieving an accuracy of 67.35%, the LP-SVD approach outperformed the other approaches by large margins (25% compared with DCT and 6 % compared with AAR-based methods). To further improve the discriminatory capability of the extracted features and reduce the computational complexity, we enlarged the extracted feature subset by incorporating two extra features, namely, Q- and the Hotelling's $T^{2}$ statistics of the transformed EEG and introduced a new EEG channel selection method. The performance of the EEG classification based on the expanded feature set and channel selection method was compared with that of a number of the state-of-the-art classification methods previously reported with the BCI IIIa competition data set. Our method came second with an average accuracy of 81.38%.
format	Article
author	Baali, Hamza Khorshidtalab, Aida Mesbah, Mustafa Salami, Momoh Jimoh Eyiomika
author_facet	Baali, Hamza Khorshidtalab, Aida Mesbah, Mustafa Salami, Momoh Jimoh Eyiomika
author_sort	Baali, Hamza
title	A transform-based feature extraction approach for motor imagery tasks classification
title_short	A transform-based feature extraction approach for motor imagery tasks classification
title_full	A transform-based feature extraction approach for motor imagery tasks classification
title_fullStr	A transform-based feature extraction approach for motor imagery tasks classification
title_full_unstemmed	A transform-based feature extraction approach for motor imagery tasks classification
title_sort	transform-based feature extraction approach for motor imagery tasks classification
publisher	IEEE
publishDate	2015
url	http://irep.iium.edu.my/46812/ http://irep.iium.edu.my/46812/ http://irep.iium.edu.my/46812/ http://irep.iium.edu.my/46812/1/07299634_%281%29.pdf
first_indexed	2023-09-18T21:06:37Z
last_indexed	2023-09-18T21:06:37Z
_version_	1777410975860785152

A transform-based feature extraction approach for motor imagery tasks classification

Similar Items