PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Optimizing Spectral Filters for Single Trial EEG Classification
Ryota Tomioka, Guido Dornhege, Guido Nolte, Kazuyuki Aihara and Klaus-Robert Müller
In: DAGM, LNCS 4174, 12-14 Sep 2006, Berlin, Germany.

Abstract

We propose a novel spectral filter optimization algorithm for single trial ElectroEncephaloGraphy (EEG) classification problem. The algorithm is designed to improve the classification accuracy of Common Spatial Pattern (CSP) based classifiers. The algorithm is based on a simple statistical criterion, and allows the user to incorporate any prior information one has about the signal. We show that with a different preprocessing, how a prior knowledge can drastically improve the classification or be completely betrayed. We also show a generalization of the CSP algorithm so that the CSP spatial projection can be recalculated after the optimization of the spectral filter. This leads to an iterative procedure of spectral and spatial filter update that further improves the classification accuracy, not only by imposing a spectral filter but also by choosing a better spatial projection.

EPrint Type:Conference or Workshop Item (Paper)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Brain Computer Interfaces
ID Code:2279
Deposited By:Guido Dornhege
Deposited On:21 October 2006