Wearable Computing for Physical Therapy

The project will develop a novel wearable computing system for use in physical therapy and rehabilitation exercises. Often patients undergoing physical therapy are prescribed a set of stretches and exercises to strengthen an injured muscle group. While outside of the doctor’s office, the physical therapist cannot observe if the activities are being performed correctly or if at all. Wearable computing solutions have been developed that will allow human activities to be recognized. However, current approaches are limited to specific arm and leg motions and not the broader set of activities that would be required in a general tool usable by medical practitioners. The project would leverage existing machine learning and human-computer interaction techniques to develop a custom wearable computing system tailored to rehabilitation exercises of a particular patient. This system would create positive patient outcomes by providing feedback to the patient and the physical therapist about the quantity and quality of therapy activities performed.

Publications from this work:

Sophia Cronin, Tyler Webster, Jason Forsyth “Student Research Short: Wearable Computing for Physical Rehabilitation”, Poster Presentation at 2021 ACM Capital Region Celebration of Women in Computing (CAPWIC) (refereed as abstract) link

Sanarea Ali, “Wearable Computing for Physical Rehabilitation”, at the 2019 ACM Capital Region Celebration of Women in Computing Conference (CAPWIC), March 23rd, 2019 link

Funding for the work:

JMU CISE Mini Grant: J. Forsyth,“Prototyping Wearables for Physical Therapy”. $700. Spring 2021.

JMU CISE Faculty Development Grant: J. Forsyth (PI), M. Stewart (Co-PI), “A Wearable Computing System to Increase Access to Healthcare and Patient Outcomes in Physical Rehabilitation Exercises”. $2,500. Spring 2019.

Jason Forsyth
Associate Professor of Engineering

Jason Forsyth is an Associate Professor of Engineering at James Madison University. His major research interests are in wearable/ubiquitous computing and engineering education. His current research interests focus on on-body human activity recognition and interactive machine learning for physical therapy patients and practitioners to increase exercise adherence and clinical evaluation.


According to the National Library of Medicine, more than 50% of the US population is affected by musculoskeletal impairments, making it …

With the growth of wearable devices, professional athletes are frequently monitored during training and competition to assess athletic …