Development of an EEG-based Brain-Controlled System for a Virtual Prosthetic Hand

Document Type

Conference Proceeding

Publication Date



Meant to improve the overall quality of life for those with physical or motor impairments, this paper explores the use of EEG and its potential in controlling a prosthetic hand. EEG signal acquisition is centered on oscillatory features through the sensory motor rhythm which can be obtained through motor-imagery (MI). The EEGNet, a convolutional neural network, is used for feature extraction and signal classification of five motor-imagery classes of a hand. A reinforced model through a transfer learning approach deemed to have the best cross-validation accuracy. A real-time debugging module for the virtual hand was implemented using MuJoCo HAPTIX.