An Adaptive Closed-Loop Robotic Exoskeleton for Upper Extremity Motor Rehabilitation

University Of Rhode Island, Kingston RI

Investigators

Abstract

Upper limb disability in individuals with stroke has devastating impacts on their quality of life over their lifespan. Each year, approximately 800,000 new stroke cases are reported in the United States alone. The restoration of arm extremity and hand dexterity is the highest priority among this population. In recent years, assistive robots and rehabilitation exoskeleton platforms showed promising results in motor training and recovery of upper limb function. However, the currently available robotic and exoskeleton platforms are not affordable and require technicians and large clinical space for operation. Additionally, the current robotic control algorithms are not efficient in persuading and engaging the patient into the loop of training. This award supports research to develop innovative adaptive algorithms embedded in an innovative and portable exoskeleton platform for arm extremity training in stroke patients. The project’s affordable, user-friendly robotic interface has the potential to relieve the burden on healthcare workers and accelerate the scalability of the overall system from doctor’s office-based training and testing to home-use devices to fulfill patients’ rehabilitation needs. This project's goals will be accomplished through three research thrusts: 1) developing a multimodal, wearable exoskeleton actuated using a haptic forcefield for upper extremity training; 2) leveraging the shared control theory to develop a closed-loop adaptive assistive strategy; and 3) validating the proposed rehabilitative platform on stroke patients with upper extremity impairment. In this project, a new, multimodal, and portable planar robotic training platform with a convenient user-centered design will be developed to overcome the translational barriers and assist the recovery of arm extremity in stroke patients. By leveraging the shared control theory and using a novel, adaptive, closed-loop Kalman filter algorithm, an adaptive and intention-driven rehabilitative algorithm will be developed. The patient’s multimodal biomarkers will be incorporated in the form of an adaptive, assist-as-needed, closed-loop algorithm to accelerate the recovery and remedy of the cortical plasticity. This research work will not only advance the fundamental understanding of the role of multimodal cortico-muscular activities in the planning and execution of arm extremity but also produce new adaptive rehabilitative algorithms to accelerate the motor recovery in the affected stroke population. This project is jointly funded by the Disabilities and Rehabilitation Engineering Program (DARE) and the Established Program to Stimulate Competitive Research (EPSCoR). This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.