Enhanced active segment selection for single-trial EEG classification

Wei Yen Hsu

Research output: Contribution to journalArticle

21 Citations (Scopus)

Abstract

In this study, an electroencephalogram (EEG) analysis system is proposed for single-trial classification of both motor imagery (MI) and finger-lifting EEG data. Applying event-related brain potential (ERP) data acquired from the sensorimotor cortices, the system mainly consists of three procedures; enhanced active segment selection, feature extraction, and classification. In addition to the original use of continuous wavelet transform (CWT) and Student 2-sample t statistics, the two-dimensional (2D) anisotropic Gaussian filter further refines the selection of active segments. The multiresolution fractal features are then extracted from wavelet data by using proposed modified fractal dimension. Finally, the support vector machine (SVM) is used for classification. Compared to original active segment selection, with several popular features and classifier on both the MI and finger-lifting data from 2 data sets, the results indicate that the proposed method is promising in EEG classification.

Original languageEnglish
Pages (from-to)87-96
Number of pages10
JournalClinical EEG and Neuroscience
Volume43
Issue number2
DOIs
Publication statusPublished - Apr 2012
Externally publishedYes

Fingerprint

Electroencephalography
Fractals
Imagery (Psychotherapy)
Fingers
Wavelet Analysis
Evoked Potentials
Students
Brain

Keywords

  • active segment selection
  • brain-computer interface (BCI)
  • electroencephalogram (EEG)
  • modified fractal dimension
  • support vector machine (SVM)
  • wavelet transform

ASJC Scopus subject areas

  • Clinical Neurology
  • Neurology

Cite this

Enhanced active segment selection for single-trial EEG classification. / Hsu, Wei Yen.

In: Clinical EEG and Neuroscience, Vol. 43, No. 2, 04.2012, p. 87-96.

Research output: Contribution to journalArticle

@article{00c1d754cc324498ad657fe84ce2e036,
title = "Enhanced active segment selection for single-trial EEG classification",
abstract = "In this study, an electroencephalogram (EEG) analysis system is proposed for single-trial classification of both motor imagery (MI) and finger-lifting EEG data. Applying event-related brain potential (ERP) data acquired from the sensorimotor cortices, the system mainly consists of three procedures; enhanced active segment selection, feature extraction, and classification. In addition to the original use of continuous wavelet transform (CWT) and Student 2-sample t statistics, the two-dimensional (2D) anisotropic Gaussian filter further refines the selection of active segments. The multiresolution fractal features are then extracted from wavelet data by using proposed modified fractal dimension. Finally, the support vector machine (SVM) is used for classification. Compared to original active segment selection, with several popular features and classifier on both the MI and finger-lifting data from 2 data sets, the results indicate that the proposed method is promising in EEG classification.",
keywords = "active segment selection, brain-computer interface (BCI), electroencephalogram (EEG), modified fractal dimension, support vector machine (SVM), wavelet transform",
author = "Hsu, {Wei Yen}",
year = "2012",
month = "4",
doi = "10.1177/1550059412445051",
language = "English",
volume = "43",
pages = "87--96",
journal = "Clinical EEG and Neuroscience",
issn = "1550-0594",
publisher = "EEG and Clinical Neuroscience Society (ECNS)",
number = "2",

}

TY - JOUR

T1 - Enhanced active segment selection for single-trial EEG classification

AU - Hsu, Wei Yen

PY - 2012/4

Y1 - 2012/4

N2 - In this study, an electroencephalogram (EEG) analysis system is proposed for single-trial classification of both motor imagery (MI) and finger-lifting EEG data. Applying event-related brain potential (ERP) data acquired from the sensorimotor cortices, the system mainly consists of three procedures; enhanced active segment selection, feature extraction, and classification. In addition to the original use of continuous wavelet transform (CWT) and Student 2-sample t statistics, the two-dimensional (2D) anisotropic Gaussian filter further refines the selection of active segments. The multiresolution fractal features are then extracted from wavelet data by using proposed modified fractal dimension. Finally, the support vector machine (SVM) is used for classification. Compared to original active segment selection, with several popular features and classifier on both the MI and finger-lifting data from 2 data sets, the results indicate that the proposed method is promising in EEG classification.

AB - In this study, an electroencephalogram (EEG) analysis system is proposed for single-trial classification of both motor imagery (MI) and finger-lifting EEG data. Applying event-related brain potential (ERP) data acquired from the sensorimotor cortices, the system mainly consists of three procedures; enhanced active segment selection, feature extraction, and classification. In addition to the original use of continuous wavelet transform (CWT) and Student 2-sample t statistics, the two-dimensional (2D) anisotropic Gaussian filter further refines the selection of active segments. The multiresolution fractal features are then extracted from wavelet data by using proposed modified fractal dimension. Finally, the support vector machine (SVM) is used for classification. Compared to original active segment selection, with several popular features and classifier on both the MI and finger-lifting data from 2 data sets, the results indicate that the proposed method is promising in EEG classification.

KW - active segment selection

KW - brain-computer interface (BCI)

KW - electroencephalogram (EEG)

KW - modified fractal dimension

KW - support vector machine (SVM)

KW - wavelet transform

UR - http://www.scopus.com/inward/record.url?scp=84865769776&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84865769776&partnerID=8YFLogxK

U2 - 10.1177/1550059412445051

DO - 10.1177/1550059412445051

M3 - Article

C2 - 22715494

AN - SCOPUS:84865769776

VL - 43

SP - 87

EP - 96

JO - Clinical EEG and Neuroscience

JF - Clinical EEG and Neuroscience

SN - 1550-0594

IS - 2

ER -