THE ROLE OF ARTIFICIAL INTELLIGENCE IN PHYSICAL THERAPY AND REHABILITATION: A SCOPING REVIEW

Abstract

Background: Modern technology of Artificial intelligence (AI) has rapidly emerged as a transformative force in healthcare, with growing applications in rehabilitation and physical therapy. From automated movement analysis to predictive recovery modeling, AI promises to enhance efficiency, accuracy, and personalization of care. However, evidence is scattered, and implementation remains limited.

Objective: This scoping review aimed to map the evidence on AI applications in physical therapy and rehabilitation, summarize reported benefits and challenges, and identify knowledge gaps to guide future research and policy.

Methods: Following Arksey & O’Malley’s scoping review framework and PRISMA-ScR guidelines, we searched PubMed, PEDro and Cochrane Library, eligible studies were those in which participants were human, focusing on AI applications in physical therapy and rehabilitation. Data were charted on AI type, clinical domain, outcomes, and barriers.

Results: Forty-seven eligible studies were identified. Most research focused on neurologic rehabilitation (stroke, Parkinson’s, traumatic brain injury) and musculoskeletal disorders. Common AI approaches included machine learning, computer vision, wearable analytics, and AI-driven robotics. Benefits included automated assessment, personalized planning, improved adherence, and remote monitoring. Reported barriers included data quality, lack of interoperability, “black-box” concerns, high costs, and limited clinician training. Few large, multi-center clinical trials or cost-effectiveness studies were found.

Conclusions: AI holds strong potential in physical therapy and rehabilitation, but real-world integration is constrained by educational, organizational, and infrastructural barriers. Future work should prioritize pragmatic trials, explainable AI models, workforce training, and ethical governance frameworks.