Understanding Functional Brain Activation using Source Localization of EEG Signals in Motor Imagery Tasks

Abstract

EEG based Brain Computer Interfaces (BCIs) have been extensively researched upon to facilitate healthcare solutions because of their cost-effectiveness, portability, ease of use, and non-invasiveness. Among various healthcare technologies that can be designed using BCIs, assistive technologies such as orthotics, prosthetics and rehabilitative training devices are crucial as they can aid people with motor disabilities. The pre-requisite for developing accurate BCIs for such technologies require neuro-feedback corresponding to movement perception, imagery and/or execution. At the same time, it is crucial to identify whether the EEG signals are localized to the areas of the brain that are known to activate during motor imagery (MI). Furthermore, source localization methods can be used to classify motor imagery tasks more efficiently than scalp-level, as the features are extracted from source-level. In this article, the functional brain areas corresponding to motor imagery for four classes (right hand, left hand, both feet and tongue) have been identified using source localization. Utilization of source localized data for MI based BCIs can facilitate more effective classification of motor imagery.