Research of Feature Extraction Method for Stroke Patients´ Surface Electromyography

Surface electromyography is a one-dimensional time series signal of neuromuscular system recorded from skin surface. It can reflect the states of muscle activity and muscle function accurately. All the subjects had to perform dynamic contraction for stroke´s knee flexion and extension in experiment. The surface electromyography were collected by surface electrodes and then processed by linear time and frequency-domain method. SEMG characteristics extraction has been done and an eigenvector space of mode recognition was built, and lies the theoretical and technical foundation for stroke patients´ rehabilitation training.