Pseudo-online classification of mental tasks

Abstract

This paper presents the classification of three mental tasks, using the electroencephalographic signal and simulating a real-time process. Three types of classifiers are compared: k-nearest neighbors, Linear Discriminant Analysis and feed-forward backpropagation Artificial Neural Networks. The mental tasks are the imagination of right or left hand movements and generation of words beginning with the same random letter. The real-time simulation uses the sliding window technique, and the feature extraction uses the Power Spectral Density. A reclassification model is proposed to stabilize the classifier, and the Sammon map is used to visualize the class separation. Finally, it is expected that the proposed method can be implemented in a brain-computer interface associated with a robotic wheelchair.