AI-Enabled Precision Musculoskeletal Rehabilitation: Integrating Digital Biomarkers, Personalized Exercise Therapy, and Functional Recovery

Abstract

Background: Musculoskeletal (MSK) rehabilitation traditionally relies on episodic, protocol-driven care, often resulting in suboptimal functional recovery and poor adherence to home exercise programs (HEP). Artificial intelligence (AI) and digital biomarkers offer a pathway to precision rehabilitation. This study evaluated the clinical efficacy of an AI-enabled precision rehabilitation (AI-PR) system integrating continuous digital biomarkers and personalized exercise therapy compared to standard of care (SoC) in post-operative orthopedic patients. Methods: A multi-center, parallel-group, randomized controlled trial was conducted across four tertiary hospitals in Karachi and Islamabad, Pakistan. We enrolled 280 adults (aged 18–60) who underwent primary ACL reconstruction or total knee arthroplasty. Participants were randomized 1:1 to AI-PR (utilizing wearable inertial sensors, smartphone-based markerless pose estimation, and AI-driven dynamic exercise dosing) or SoC (static paper-based HEP and bi-weekly clinic visits). The primary outcomes were time to Return to Work/Normal Daily Activities (RTW/NDL) and Lower Extremity Functional Scale (LEFS) scores at 24 weeks. Secondary outcomes included HEP adherence, kinematic symmetry, and ecological daily step counts. Results: Of the 280 enrolled patients, 264 completed the 24-week follow-up (AI-PR, n=133; SoC, n=131). The AI-PR group achieved RTW/NDL significantly faster than the SoC group (median 48 vs. 67 days, p < 0.001). At 24 weeks, the AI-PR group demonstrated superior functional recovery (LEFS: 68.5 ± 6.4 vs. 58.1 ± 7.8, p < 0.001), higher HEP adherence (88.4% vs. 54.1%, p < 0.001), and greater ecological mobility (8,450 vs. 5,920 daily steps, p < 0.001). Outcomes were consistent across both the Karachi and Islamabad cohorts. Conclusion: AI-enabled precision rehabilitation significantly accelerates functional recovery, enhances exercise adherence, and improves real-world mobility compared to standard care. By leveraging digital biomarkers for dynamic exercise personalization, AI-PR provides a scalable, ecologically valid framework for MSK rehabilitation that is highly effective across diverse urban environments in developing healthcare systems.