Robotic Retraining of Hand Function Following Neurological Injury
Most stroke survivors have some upper extremity impairment. Because the hand is crucial to many activities of daily living, limited hand function can have wide-ranging effects on one's independence post-stroke. Robotic rehabilitation is emerging as a tool to increase access to and dosage of therapeutic exercises while minimizing the cost and labor burden of rehabilitative therapy. The Hand Exoskeleton Rehabilitation Robot (HEXORR) is a novel system designed to retrain hand function following stroke. This thesis examines the neurophysiological effects of HEXORR practice, compares HEXORR's Tone compensation to another assistance method, and examines the benefits of HEXORR training post-stroke.To examine the neurophysiological effects of HEXORR training, we created a bias such that one muscle is used over its antagonist throughout a training session. We then trained healthy volunteers on the device and measured physiological changes using transcranial magnetic stimulation. We found response to training varied based on the muscle used in the task rather than the muscle's role as an agonist or antagonist in the task. This information can be used to improve treatment plans for stroke patients.Oftentimes, robotic assistance is modeled after a spring attaching one's limb to a target. The further one is from the target, the larger the force produced by the robot to assist in reaching the target. In contrast, HEXORR uses Tone Assistance, a method based on the user's own hypertonicity, resulting in increased assistance as one nears the target. We compared the performance of stroke patients using these two modes and found that the Tone mode encourages the user to be more engaged in the training and enables more natural torque patterns to produce these movements.Finally, we conducted a training study with 8 stroke patients in which they trained with HEXORR for 18 sessions. We measured their progress by comparing clinical measures of impairment before and after training and at follow-up. Overall, we found subjects stratified into lower and higher impairment groups which were related to outcome. Those with higher impairment improved only in range of motion and grip strength whereas those with lower impairment improved on a range of clinical scales.
StatsViewed 31 times
Downloaded 3 times