Artificial intelligence in physical rehabilitation: A systematic review

Area of Science:

• Gerontology and Rehabilitation Science
• Biomedical Engineering
• Health Informatics

Background:

• Physical disabilities increase with age, making rehabilitation crucial for maintaining independence.
• Current rehabilitation services face challenges in availability and accessibility, limiting their impact.
• The role of Artificial Intelligence (AI) in enhancing physical rehabilitation remains largely unexplored.

Approach:

• A systematic review was conducted on AI-supported physical rehabilitation technologies tested in clinical settings.
• Searches were performed across multiple major scientific databases (MEDLINE, EMBASE, CINAHL, Web of Science, CIRRIE, OpenGrey).
• The review focused on AI technology availability, clinical effects, and implementation barriers/facilitators.

Key Points:

• 28 projects involving AI in rehabilitation were identified, categorized into app-based systems, robotic devices, gaming systems, and wearables.
• Five randomized controlled trials (RCTs) showed inconsistent clinical effects on physical function, activity, pain, and quality of life.
• Implementation barriers included technology literacy and user fatigue, while enablers included improved access and cost reduction.

Conclusions:

• AI application in physical rehabilitation is an emerging field with potential benefits.
• Current evidence on clinical effectiveness is limited and requires more rigorous investigation.
• Future research must prioritize robust clinical evaluations and real-world implementation studies.