ARTIFICIAL INTELLIGENCE IN COMMUNITY BASED REHABILITATION: A SYSTEMATIC REVIEW

Abstract

Background: Community-Based Rehabilitation (CBR) is a strategy to promote inclusion, independence, and participation for people with disabilities, particularly in low-resource settings. In recent years, artificial intelligence (AI) has been increasingly used in clinical rehabilitation; however, its integration into community contexts remains limited. Objective: This systematic review aimed to identify and synthesize recent evidence on the use of AI technologies within CBR. Methods: A systematic search was conducted in PubMed, Scopus, and ScienceDirect for articles published from January 2020 to July 2025. Eligible studies included empirical research applying AI in CBR contexts. Two reviewers independently screened, extracted data, and assessed risk of bias using RoB2, ROBINS-I, PROBAST, and QUADAS-2. Results: From 842 identified records, 10 studies met inclusion criteria. Applications of AI in CBR were grouped into prediction and screening, home-based or remote rehabilitation, patient empowerment, and social support. Reported benefits included improved cognition, sarcopenia reversal, frailty and depression screening, diabetes self-management, and smoking cessation. Socially assistive robots were found acceptable and useful in supporting daily activities and emotional well-being. Limitations across studies included small samples, short follow-up, limited external validation, and a focus on technologically literate populations. Conclusion: AI shows considerable potential to strengthen the accessibility, personalization, and effectiveness of CBR. Future research should focus on large-scale, long-term studies with diverse populations and explore strategies for equitable, sustainable integration of AI into community rehabilitation services.