Controlling Artificial Limb Movement System using EEG Signals

Objectives: We mainly focus the application of machine learning for artificial limb movement system using Electroencephalogram (EEG) signals. Analysis: EEG signals depict the neuronal activity happening in brain, which will be used to control the artificial limb movement system. Findings: In this paper, four classes of EEG signals were recorded from healthy subjects while performing actions such as finger open (fopen), finger close (fclose), wrist clockwise (wcw) and wrist counterclockwise (wccw) movements. The main objective of this study is to extract the statistical features from EEG signals and identify the best possible features and classify them using J48 Decision Tree algorithm. Improvements: The EEG signals are complex in nature and machine-learning approach was used to study the same. To improve the classification accuracy better feature extraction techniques might be used.