Movement Training and Post-Stroke Rehabilitation Using a Six Degree of Freedom Upper-Extremity Robotic Orthosis and Virtual Environment

Prior research has shown that high-intensity repetitive movement training is an effective method for restoring some function to patients with chronic hemiparesis resulting from a stroke. The goal of this dissertation was to develop and evaluate a robotic system for assisting in post-stroke upper-extremity rehabilitative physical therapy. The primary contributions of this research are in: the development of control software for a rehabilitation robot; the development of a virtual environment for rehabilitation; the integration between a robot controller and a virtual environment; the design and analysis of an algorithm for maintaining a challenging level of therapy; and a clinical trial testing the efficacy of functional multi-joint movement therapy (FMJMT) versus isolated individual joint movement training (IIJMT).;The robotic mechanism created to provide therapy is a 6 degree-of-freedom (DOF) pneumatically actuated exoskeleton, which included a 3 DOF shoulder, a 1 DOF elbow, and a 2 DOF module for forearm supination/pronation and wrist flexion/extension. The shoulder is driven by redundant actuation of a novel spatial parallel mechanism. The elbow, forearm, and wrist are all actuated in serial with four-bar slider-crank mechanisms.;The virtual environment consisted of several games with point-to-point reaching or continuous tracking tasks. Some of the games required squeezing a pressure sensitive device for grasping items in the games. There was also another novel environment that allowed for a coordinated movement strength assessment. An algorithm that was developed to discretely alter the game difficulty was implemented in some games to maintain a challenging level of therapy for the patients. This algorithm is also shown to detect motor improvement.;The clinical trial involved a crossover design study with 20 patients who had chronic hemiparesis after stroke. Of these, 10 patients received FMJMT for 3 one hour sessions per week for 4 weeks followed by IIJMT for the same duration. The other 10 subjects received both interventions in the opposite order. The results from the study indicated both training techniques promoted motor recovery with slightly more benefit from IIJMT suggesting that the complexity of multi-DOF robots may not be necessary for promoting upper limb recovery in hemiplegic stoke survivors.