Wearable Hand Robots for Rehabilitation and Daily Assistance
Keywords: Wearable Robots, Stroke Rehabilitation, Robot Design and Control, Soft Robotics, Machine Learning, Assistive Devices
Overview of Research
Approximately 38% of stroke survivors (~200,000 Canadians) report that their greatest motor impairment is in their hands, and 20% (~100,000 Canadians) have hypertonia in their finger flexors and weakness in their finger extensors, leading to minimal active extension and clenched non-functional hands. This prevents them from independently performing daily tasks that require grasping with the supporting hand. The goal of this project is to develop a novel and usable robotic orthosis that improves stroke survivorsí grasp performance during functional tasks in daily life.
Previously, a novel robotic orthosis was developed that applies controlled forces and motions to the stroke-affected fingers. This robot enables the passive and active biomechanics of stroke-affected fingers to be quantified, and may safely extend clenched fingers to functional positions. In addition, the robot is wearable for future use in rehabilitation clinics and throughout daily life.
The next stage is to optimize how the device functions when used independently by stroke survivors. This requires (i) algorithm development to detect usersí intent to flex and extend their fingers, and (ii) design iteration to make the system comfortable, lightweight, and miniaturized for extended use in daily life. The previously-developed robot will quantify residual finger flexion-extension ability that machine learning algorithms will use to estimate user intent and automatically trigger personalized robot assistance as needed. The biomechanics and residual ability data will be used to optimize the selection of polymer sensors, actuators and batteries to create a novel robot orthosis that is compliant, lightweight and discreet. The developed robotic orthosis will continue to be tested at the Toronto Rehabilitation Institute to evaluate the stroke-affected hand's daily-task performance and overall usability while wearing the smart soft robotic orthosis.
Current work on the project involve employing wide angle lenses to increase the area of coverage, improving robustness with respect to lighting changes, and applying state-of-the-art statistical analysis and machine learning techniques in order to reduce the burden of collecting and labelling training data during the system development process.
Post-Graduate Scholarship - Doctoral, National Sciences and Engineering Research Council (NSERC)
Trainee Award, Canadian Partnership in Stroke Recovery
Graduate Student Award in Technology & Aging, AGE-WELL NCE
Faculty of Engineering Award, University of Toronto
Alex Mihailidis, University of Toronto
Aaron Yurkewich, Toronto Rehabilitation Institute
Rosalie Wang, University of Toronto