BCI Rehabilitation System Based On Dynamic Patten Recognition Method

In recent years, stroke is becoming a more and more common dis-ease. Post-stroke sequela will damage the cranial nerve, resulting in anafect in the post-stroke patients’ daily life. The Function Electrical S-timulation(FES) rehabilitation system based on motor imagery, which cancreate a neuron circuit between post-stroke patients’ motor imagery andreal body movement, is one of the most important topic in Brain Comput-er Interface(BCI) system. While applying the system on reality, we foundhuge diferences in the motor imagery signal between post-stroke patientsand the normals. Thus we try to use dynamic pattern recognition methodand design a dynamic pattern based BCI rehabilitation system, and applythe system on the clinical experiment. We compare the EEG signal frompost-strokepatients and the normals, fnd out the diference and extractdy-namic feature that can represent the diference. We try some most widelyused dynamic pattern recognition method and get some results.