Reinforcement learning methods to automate spinal cord stimulation and improve motor recovery
Thematic Area: Health Care and Wellness
Faculty Advisor: Ashley Dalrymple, Department of Biomedical Engineering
After spinal cord injuries, individuals often experience permanent motor impairments, and regaining the ability to walk is a high priority. Electrical stimulation of the lumbosacral spinal cord using epidural spinal cord stimulation (eSCS) has shown potential to improve motor recovery by amplifying weak but voluntary limb movements. Current approaches for selecting eSCS electrodes and stimulation parameters are manual and burdensome. Grange Simpson will develop reinforcement learning methods to automatically select and update stimulation parameters over time to promote optimal muscle activity and joint movement for patient recovery. His predictive reinforcement learning framework will address an unmet need with eSCS for motor rehabilitation, improving the lives of individuals living with spinal cord injury.
Simpson began working toward his biomedical engineering Ph.D. at the University of Utah in 2023. He holds bachelor’s and master’s degrees from the same department.
