Abstract:
Stroke is a leading cause of adult death and disability, often followed by muscle weakness, loss of control and coordination in limbs and movement disorders. Consequently, stroke-surviving individuals with hemiplegia are often unable to perform simple tasks like opening and closing of their affected (unhealthy) hand, making them dependent on a caregiver for day-to-day activities. At the same time their healthy hand retains ability for normal activities. Rehabilitation aims to improve their ability to use their affected hand similar to their use of healthy hand. The degree of closure (flexion) of one’s hand can be mapped from surface Electromyogram (sEMG) signal of Flexor Carpi Radialis muscle present on the anterior side of one’s forearm. The different degrees of flexion can be classified with the help of Support Vector Machines (SVM) using the sEMG signals from the Flexor Carpi Radialis muscle. In this study we developed a proof-of-concept Virtual Reality based real-time rehabilitative system for post-stroke hand movement disorder. Our developed system uses the sEMG data obtained from the healthy hand of stroke-surviving individuals as training dataset for classifying the degree of flexion of their respective stroke-affected hand while they interact with the VR-based tasks, and triggers therapeutic electrical stimulation to be applied to the muscles of the unhealthy hand of the stroke-surviving individuals based on their performance feedback. This system can be used by the patient at home as per his convenience, with minimal dependency on a physiotherapist or a
caregiver. The preliminary results of testing and feasibility studies suggest that the hand flexion and extension skills of the participants (six able-bodied and two stroke-surviving persons) improved with repeated attempts. This indicates that our system has the potential to take a definite step towards becoming a simple, technology-assisted solution for rehabilitation of hand movement disorder.