Using NIRS as a predictor for EEG-based BCI performance

Author

Fazli, Siamac ; Mehnert, J. ; Steinbrink, J. ; Blankertz, Benjamin

Author_Institution

Dept. of Comput. Sci., Tech. Univ. Berlin, Berlin, Germany

fYear

2012

fDate

Aug. 28 2012-Sept. 1 2012

Firstpage

4911

Lastpage

4914

Abstract

Multimodal recordings of EEG and NIRS of 14 subjects are analyzed in the context of sensory-motor based Brain Computer Interface (BCI). Our findings indicate that performance fluctuations of EEG-based BCI control can be predicted by preceding Near-Infrared Spectroscopy (NIRS) activity. These NIRS-based predictions are then employed to generate new, more robust EEG-based BCI classifiers, which enhance classification significantly, while at the same time minimize performance fluctuations and thus increase the general stability of BCI performance.

Keywords

biomedical optical imaging; brain-computer interfaces; electroencephalography; fluctuations; infrared imaging; infrared spectra; medical signal processing; signal classification; EEG-based BCI classifiers; EEG-based BCI control; EEG-based BCI performance; classification enhancement; near-infrared spectroscopy; performance fluctuations; sensory-motor based brain computer interface; Brain computer interfaces; Educational institutions; Electrodes; Electroencephalography; Neuroimaging; Standards; Training; Adult; Algorithms; Brain Mapping; Brain-Computer Interfaces; Cerebral Cortex; Electroencephalography; Feedback, Sensory; Female; Humans; Male; Oxygen Consumption; Reproducibility of Results; Sensitivity and Specificity; Spectroscopy, Near-Infrared; Young Adult;

fLanguage

English

Publisher

ieee

Conference_Titel

Engineering in Medicine and Biology Society (EMBC), 2012 Annual International Conference of the IEEE

Conference_Location

San Diego, CA

ISSN

1557-170X

Print_ISBN

978-1-4244-4119-8

Electronic_ISBN

1557-170X

Type

conf

DOI

10.1109/EMBC.2012.6347095

Filename

6347095