Computational Intelligence and Neuroscience 
Volume 2007 (2007), Article ID 57180, 11 pages
doi:10.1155/2007/57180
Research Article

Classifying EEG for Brain-Computer Interface: Learning Optimal Filters for Dynamical System Features

Le Song1 and Julien Epps2

1School of Information Technologies, The University of Sydney, N.S.W. 2006, Australia
2National ICT Australia, Locked Bag 9013 Alexandria, N.S.W. 1435, Australia

Received 19 February 2007; Accepted 27 May 2007

Recommended by Andrzej Cichocki

Abstract

Classification of multichannel EEG recordings during motor imagination has been exploited successfully for brain-computer interfaces (BCI). In this paper, we consider EEG signals as the outputs of a networked dynamical system (the cortex), and exploit synchronization features from the dynamical system for classification. Herein, we also propose a new framework for learning optimal filters automatically from the data, by employing a Fisher ratio criterion. Experimental evaluations comparing the proposed dynamical system features with the CSP and the AR features reveal their competitive performance during classification. Results also show the benefits of employing the spatial and the temporal filters optimized using the proposed learning approach.