Presentation
Heuristic Evaluation and Usability Testing of an Artificial Intelligence-Powered Therapeutic Exercise System for Managing Knee Osteoarthritis in Older Adults
SessionPoster Session 2
DescriptionThis study evaluated the usability of an artificial intelligence (AI)-powered therapeutic exercise system for older adults with knee osteoarthritis (KOA) to support home-based strengthening, stretching, and balance exercises. The system integrates motion-capture technology and real-time AI feedback to emulate a physical therapist. Two human factors practitioners independently evaluated the system against established usability and user interface heuristics to identify design deficiencies. Eleven older adults with KOA (mean age ± standard deviation = 67.7 ± 3.0 years) participated in usability testing and completed tasks that simulated real-world system use. A researcher observed and documented their interactions, areas of confusion, errors, and assistance requests. After the testing, the older adults provided feedback on the system through the System Usability Scale and semi-structured interviews. Key usability issues identified included a complex camera setup, insufficiently detailed exercise demonstrations, limited exercise selection flexibility, inability to pause/rewind exercises, and poor error tolerance. The study also identified deficiencies in the graphical user interface. Based on these findings, actionable design improvements are proposed to enhance the system’s usability and guide future digital health technology design.
Contributors
Event Type
Poster
TimeWednesday, October 15th5:30pm - 6:30pm CDT
LocationRiverside East
Similar Presentations