Subspace estimation approach to P300 detection and application to Brain-Computer Interface

Author

Rivet, B. ; Souloumiac, A.

Author_Institution

CEA, LIST, Gif-sur-Yvette

fYear

2007

fDate

22-26 Aug. 2007

Firstpage

5071

Lastpage

5074

Abstract

Brain-computer interface (BCI) is a system for direct communication between brain and computer. In this work, a new unsupervised algorithm is introduced for P300 subspace estimation: the raw EEG are thus enhanced by projection on the estimated subspace. Moreover a simple scheme to detect the P300 potentials in the human EEG by dimension reduction and linear support vector machine (SVM) is proposed to build a BCI based on the P300 speller. The proposed algorithm is finally tested with dataset from the BCI Competition 2003 and gives results that compare favourably to the state of the art.

Keywords

electroencephalography; medical signal detection; medical signal processing; support vector machines; unsupervised learning; user interfaces; P300 potential detection; SVM; brain-computer interface; dimension reduction; linear support vector machine; raw human EEG; subspace estimation approach; unsupervised algorithm; Algorithm design and analysis; Application software; Brain computer interfaces; Computer interfaces; Electroencephalography; Humans; Independent component analysis; Neuromuscular; Support vector machines; Testing; Algorithms; Artificial Intelligence; Brain; Data Interpretation, Statistical; Electroencephalography; Event-Related Potentials, P300; Humans; Pattern Recognition, Automated; Reproducibility of Results; Sensitivity and Specificity; User-Computer Interface;

fLanguage

English

Publisher

ieee

Conference_Titel

Engineering in Medicine and Biology Society, 2007. EMBS 2007. 29th Annual International Conference of the IEEE

Conference_Location

Lyon

ISSN

1557-170X

Print_ISBN

978-1-4244-0787-3

Type

conf

DOI

10.1109/IEMBS.2007.4353480

Filename

4353480