Wei-Chih Lien, Attending Physician and Director, Division of Musculoskeletal Rehabilitation, Dept. of Physical Medicine and Rehabilitation

National Cheng Kung University Hospital (NCKUH), working with NCKU Industrial Design and YunTech Information Technology research teams, developed an innnovative AI-based posture detection for frailty care. This service combines wearable sensors with artificial intelligence. Wireless inertial measurement units record gait and posture signals from older adults, and the system extracts features and uses AI models to determine frailty status, offering objective, quantitative assessments. International journals published this research, which demonstrates over 90% accuracy in frailty classification and supports its clinical viability.

In clinical practice, physicians, therapists, and case managers quickly perform assessments in outpatient clinics, rehabilitation gyms, or community health centers. They upload data to inform strength training, balance exercises, and nutritional interventions. The system also connects to home-based exercise programs, allowing older adults to follow guidance for safe at-home exercise, while sensors provide feedback on posture quality.

For patients and families, this service reduces unnecessary hospital visits and offers follow-up on frailty status, thereby enhancing satisfaction and engagement in self-care. For healthcare and rehabilitation teams, AI tools automatically process gait and posture data, helping them identify high-risk individuals precisely and improve workflow and interdisciplinary communication. Facing rapid population aging, NCKUH positions this AI-based posture detection system as a cornerstone of its smart rehabilitation care strategy, linking community screening with home-based rehabilitation and highlighting its leading role in smart care in southern Taiwan.