Regions of Interest (ROI) Analysis for Upper Limbs EEG Neuroimaging Schemes

The rapid advancement in machine learning and digital signal processing is the reason for many researchers to develop brain-computer interface (BCI) systems. Recently, Electroencephalography (EEG) is the most widely used brain signals for many BCI systems because of recent EEG devices. These devices are easy to use, low costs, versatile and portable. Most popular EEG applications are controlling for artificial devices and virtual objects, rehabilitation for disabled persons, and computer displays in a real-time system. Since EEG does not collect the activity of single neurons, it detects the signals when populations of neurons are active at the same time and it made volume conduction problem. Thus the event-specific cortex regions and robust features for high classification accuracy are still challenges. This paper analyzed EEG signals over source-based (ROI) direction and the SVM classifier is applied to measure the accuracy of the proposed ROI. This work combined preprocessing, segmentation, Source Imaging, ROI analysis, feature extraction, and classification. Defined ROI is mainly focused on upper limb Motor Imagery EEG signals. The classification results showed that the defined ROI is reliable for MI EEG data.