Motor Imagery Data Classification for BCI Application Using Wavelet Packet Feature Extraction

The noninvasive brain imaging modalities have provided us an extraordinary means for monitoring the working brain. Among these modalities, Electroencephalography (EEG) is the most widely used technique for measuring the brain signals under different tasks, due to its mobility, low cost, and high temporal resolution. In this paper we investigate the use of EEG signals in brain-computer interface (BCI) systems. We present a novel method of wavelet packet-based feature extraction and classification of motor imagery BCI data. The prominent discriminant features from a redundant wavelet feature set is selected using the receiver operating characteristic (ROC) curve and fisher distance criterion. The BCI competition 2003 data set Ib is used to evaluate a number of classification algorithms. The results indicate that ROC is able to produce better classification accuracy as compared with that from the fisher distance criterion.