Iteratively Learning Electromyography (EMG)-based Functional Electrical Stimulation (FES) for Stroke Rehabilitation

Lupu O., M. Madaschi, T. Seel, A. Cologni, F. Previdi, T. Schauer

Rehabilitation after stroke can be improved by EMG­proportional FES. While the muscle is being stimulated
proportional to its detected residual voluntary activation, the patient practices to reach pre­defined joint angles
iteratively. Often the patient has bad control over the movement and oscillations occur. A potential reason is the
nonlinear static EMG­angle relation resulting from EMG­proportional stimulation. This contribution describes the
development of an improved EMG­based FES system that iteratively learns the nonlinear EMG­FES relation that leads
to a linear relation between EMG and joint angle.