Intelligent classifier for a direct neural interface

Abstract

Brain Computer Interfacing (BCI) is one of the new emerging technologies that signify a great potential for the physically disabled. It concentrates on developing new augmentative communication and control technologies for those with severe neuromuscular disorders. It can be used to provide a means of communication for patients suffering from neurological diseases hindering normal communication due to loss of motor function. B-C interfaces can be used to allow locked-in persons to be able to use their minds to control a computer, which can in turn control any device or system. This area of research was initiated in that devices were controlled by such things that were viewable, like blinking of eyes, jaw clench etc. Electroencephalogram (EEG) is the foundation of current BCI research. It has been experimentally shown that mental activities (mathematical calculations, motion imagination) give rise to distinct patterns in the EEG signal obtained from certain areas of the scalp. The project is to design a system that can classify EEG signals related to motion, which can in turn be fed to a mechanical system to produce movement in the desired direction. This involves creating a classifier to identify the class of the EEG signals. Neural Networks is chosen as the classification technique for this project.