Brain-Computer Interface controlled functional electrical stimulation system for paralyzed arm

Restoration of upper limb movements for stroke recovery patients is an essential keystone in rehabilitative practices. The functional electrical stimulation (FES) systems had proved its success in restoring the movement in order to increase the ability to perform the daily living activities. In this study, we propose a Brain Computer Interface (BCI) system that controls the FES. The system is formed of a data acquisition system using ANT headset, signal preprocessing followed by a feature extraction module, and a classification module for the binary movements of the arm. The performance of both the online and offline learning has been studied for different combinations of features and classifiers. The wavelet and Fourier transform based features combined with the Linear Discriminant Analysis (LDA) for the classification task has proved its superiority in performance to reach 91.43% and 0.8257 for accuracy and mutual information (MI) respectively. Transcutaneous electrical stimulation (TENS) was selected for the FES module since it was optimum for our application.