Predicted Gait Speed Recovery and Patient-Reported Pain in Patients Requiring Additional Physical Therapy Following Hip and Knee Arthroplasty

Description

The presentation discusses advancements in wearable technology and mobile applications for monitoring and analyzing patient gait recovery following surgery, particularly in musculoskeletal health. Employees from Zimmer Biomet introduce their machine learning model, "Walk," which identifies gait exceptions based on mobility data collected from smartphones and wearables. The model predicts patients who may experience low gait speed at 90 days post-operation and flags those who might recover slower than anticipated or suffer from high pain levels, thereby allowing for timely clinical interventions. The study leverages data from a smartphone care management platform, trained using machine learning techniques, to enhance the understanding of gait as a vital sign and the implications for rehabilitation outcomes. Key findings highlight a correlation between flagged patients and the need for additional physical therapy. The presentation concludes with future research directions aimed at identifying recovery signatures associated with different clinical outcomes.

DOI: 10.1302/3114-251041

Specialties

Conferences