P300-based Speller Brain-Computer Interface

Abstract

Millions of people with motor disabilities so severe that they cannot communicate with their families. In spite of their motor disabilities, sensory and cognitive functions are usually still enabled. For instance, people with spinal cord injuries or amyotrophic lateral sclerosis (ALS), also called Lou Gehrig's disease. For those people brain-computer interface (BCI) may be the only hope. BCI is a system that conveys messages and commands directly from the human brain to a computer. The described BCI system in this paper is based on the P300 wave. The P300 is a positive peak of an event-related potential (ERP) that occurs 300 ms after a stimulus in the electroencephalography (EEG) signals. One of the best-known and most widely used P300 applications is the P300 speller designed by Farwell-Donchin in 1988. In this project the used P300 paradigm is constructed as a 6×6 matrix of letters and numbers is displayed, and the subject focuses on a target character while rows and columns of characters flashing. Through detection of P300 for one row and one column, the target character can be identified. EEG recordings dataset of four subjects was used, with each dataset consists of two different sessions data. In each session, the user spelt a total of (33, 36 characters including spaces) for train and test phases, respectively. The EEG data passes through different stages those are resampling, preprocessing, feature extraction, classification, and performance evaluation. The used classifier is a regularized linear discriminant analysis (RLDA) classifier which is a type of linear discriminant analysis(LDA). Typical LDA is not likely to be adequate because of the degradation of classification accuracy due to the high-dimensionality, in other words, the number of features is greater than the number of samples. The classification results of ERP paradigm resulted by an average classification accuracy of 97.57% across the four subjects and average information transfer rates (ITRs) of 21.6 bits/min and 21.4 bits/min in the first and second sessions, respectively. Finally, with this speller we tried to use technology in a way that benefits humanity generally and medicine particularly, and improves patient's life.