Conventional therapy to improve upper limb function following stroke is not effective. Only 5% of people who survive a stroke but have severe paralysis regain upper limb function. No conventional therapy is better than another, but intensity has been shown to be important. During the last decade there has been growing evidence for the effectiveness of technologies such as rehabilitation robots and electrical stimulation, to provide an enriched training environment for recovery of movement post-stroke.
In particular, use of functional electrical stimulation (FES) is motivated by a growing body of clinical evidence, and theoretical support from neurophysiology and motor learning research, which shows that its therapeutic benefit is maximised when it is applied co-incidently with a patient's own voluntary intention. A hypothesis has been proposed that explains why the increased degree of functional recovery is closely related to the accuracy with which the stimulation assists the subject's own voluntary completion of a task.
Iterative Learning Control (ILC) has been shown to be highly effective when applied to stroke rehabilitation. In recent cross disciplinary research at the University of Southampton, FES was applied to generate torque about the elbow joint, and ILC was used to update the stimulation level to assist patient's completion of a planar reaching task. To enable accurate performance, dynamic models of the arm were developed, together with model-based ILC schemes. When used in clinical trials, statistically significant results across a range of outcome measures showed that impairment in arm function reduced over the course of only 18 treatment sessions, thereby establishing the effectiveness of the approach. However improvement in motor function was only significant across tasks similar to those trained during treatment.
To maximise its potential for rehabilitation, a system is developed in this project which extends the technology to assist unconstrained 3D arm movements using FES applied to multiple muscles. This involves substantial extension to the underlying dynamic model of the system, and to the ILC schemes used to provide the precise tracking control required. This system includes a mechanical robotic unweighing system used to support the patient's arm, FES hardware, control and user software, and custom-made virtual reality software. Trials with unimpaired participants supplying no voluntary contribution confirm the efficacy of the system, and its ability to produce accurate tracking over a range of 3D tasks. Trials will shortly commence with stroke patients.