Implementation of EEG Approach for Detection of Upper Limb Movement

Electroencephalography (EEG) signals are one of the robust electrophysiological monitoring methods to detect and record the electrical activity of the brain. This paper presents the detection of movement by implementing an EEG approach whilst performing seven different tasks that cover up full range of motion (ROM) from extension to flexion. EEG signals are captured by using Emotiv Epoc+ which are recorded by using Emotiv Xavier software. In conjunction, we also use force sensor which performs as a reference to synchronize the EEG signals with the upper limb movement. EEG signals were recorded from 39 healthy subjects comprising of 28 males and 11 females. The raw data of Emotiv was load into the Matlab for filtering, preprocessing and feature extraction. The feature used in this paper are mainly based on Hjorth Parameter. Our experimental result shows that, Area Under Curve (AUC) performance increases with increases in window size. The F1 score can be increased by reducing the gap between the class imbalance. Due to class imbalance, our results are not promising. The classification accuracy can be improved based on the performance of the feature extraction. The detection movement by using EEG signals is in future will be used to assist disable people with activities of daily living such as for post stroke rehabilitation.