Detecting Arm Movements from EEG-signal with Machine Learning

Abstract

The integration of brain-computer interface (BCI) in human-computer interaction has started in a new time full of innovation and possibilities. BCI has overcome the limitations of traditional input devices and has provided direct communication between the human brain and external systems. Especially electroencephalogram (EEG)-based biometrics (BCI) have become a promising method for seamless data exchange between the human brain and applications. This paper explores how to use convolutional neural networks (CNN) and long short-term memory (LSTM) as a basic mechanism for EEG signal description, preprocessing of EEG data using ICA and MNE, and analysis to explore subtle differences in EEG signals of left and right arm movements and hypothesize the application of this technology to the steering system of the future self-propelled wheelchair.