Brain signals: feature extraction and classification using rough set methods

Authors:
Reza Fazel-Rezai;Sheela Ramanna
Affiliations:
Department of Electrical and Computer Engineering, University of Manitoba, Winnipeg, Manitoba, Canada;Department of Applied Computer Science, University of Winnipeg, Winnipeg, Manitoba, Canada
Venue:
RSFDGrC'05 Proceedings of the 10th international conference on Rough Sets, Fuzzy Sets, Data Mining, and Granular Computing - Volume Part II
Year:
2005

Citing 1
Cited 1

A New Version of Rough Set Exploration System

TSCTC '02 Proceedings of the Third International Conference on Rough Sets and Current Trends in Computing

P300 wave detection based on rough sets

Transactions on Rough Sets V

Quantified Score

Hi-index	0.00

Visualization

Abstract

A brain computer interface (BCI) makes it possible to monitor conscious brain electrical activity, via electroencephalogram (EEG) signals, and detecting characteristics of brain signal patterns, via digital signal processing algorithms. Event Related Potentials (ERPs) are measures that reflect the responses of the brain to events in the external or internal environment of the organism. P300 is the most important and the most studied component of the ERP. In this paper, a new method for P300 wave detection is introduced. It consists of two components: feature extraction and classification. For feature extraction, Mexican hat wavelet coefficients provide robust features when averaged over different scales. Classification has been carried out using rough set based methods. The overall results show that P300 wave detection can be performed using only five EEG channels. This in turn reduces the computational time compared to the averaging method that uses more channels.