Evaluation of Features for Electrode Location Robustness in Brain-Computer Interface (BCI)

Abstract

A brain-computer interface (BCI) is a system that provides an output channel for the brain and does not depend on peripheral muscles and nerves. In other words, a BCI system interprets a brain activity as simple commands and transforms brain activities into prescribed actions within its applications, i.e. the control of a wheelchair or a robot arm. The electroencephalogram (EEG) is one of non-invasive methods to record brain activities and it records electrical activities generated by cerebral cortex nerve cells from the human scalp with electrodes. EEG-based brain-computer interface has been studied actively for its a great potential for diagnoses of brain neuro-degenerative diseases and advantages such as completely non-invasive procedure that can be applied repeatedly to patients, normal adults, and children with virtually no risk or limitation and its reasonable cost. Diverse types of electrical brain activities have been used to realize EEG-based BCI systems, e.g., mu rhythm, slow cortical potential, event-related p300, and steady-state visual evoked potential. Among these activities, the one most widely used to monitor brain activities for BCI applications has been the mu (_) rhythm, which is related to motor actions. Motor imagery (MI) is the state during which the representation of a specific motor action is internally reactivated within the working memory without any overt motor output and that is governed by the principles of motor control. MI can modulate mu rhythm activities in the sensorimotor cortex without any physical movements of the body.