Distinguishing Between Left and Right Finger Movement from EEG using SVM

Author

Shoker, Leor ; Sanei, Saeid ; Sumich, Alex

Author_Institution

Centre of Digital Signal Process., Cardiff Univ.

fYear

2006

Firstpage

5420

Lastpage

5423

Abstract

A hybrid BSS-SVM method for distinguishing between left and right finger movements from the electroencephalogram (EEG) has been developed. Support vector machines (SVM) is used to effectively classify the extracted features incorporating blind source separation (BSS) and directed transfer functions (DTF). This is the basis for a brain computer interface (BCI). We analyzed 200 trials of 64 electrode EEG data from which we trained the classifier and tested our system. We demonstrated that by classification of such appropriate features we can reliably distinguish between left and right finger movements

Keywords

biomechanics; biomedical electrodes; blind source separation; electroencephalography; feature extraction; medical signal processing; signal classification; support vector machines; EEG; SVM; blind source separation; brain computer interface; directed transfer functions; electrode; electroencephalogram; extracted feature classification; left finger movement; right finger movement; support vector machines; Blind source separation; Brain computer interfaces; Data mining; Electroencephalography; Feature extraction; Fingers; Source separation; Support vector machine classification; Support vector machines; Transfer functions;

fLanguage

English

Publisher

ieee

Conference_Titel

Engineering in Medicine and Biology Society, 2005. IEEE-EMBS 2005. 27th Annual International Conference of the

Conference_Location

Shanghai

Print_ISBN

0-7803-8741-4

Type

conf

DOI

10.1109/IEMBS.2005.1615708

Filename

1615708