Integrating artificial intelligence and gamification in rehabilitation: A scoping review

The integration of artificial intelligence (AI) and gamification represents a promising direction in rehabilitation, enabling personalized training, real-time feedback, and enhanced patient engagement. However, research in this area remains fragmented, necessitating a structured synthesis. This scoping review analyzed 20 studies published over the past two decades and indexed in PubMed, Cochrane Library, Web of Science, and IEEE Xplore. Stroke rehabilitation was the most frequently addressed condition (9 studies), followed by motor dysfunction (2), rheumatoid arthritis (2), cognitive impairment (3), phantom limb pain (2), ADHD (1), and multiple sclerosis (1). AI techniques included machine learning (10 studies), deep learning (4), and unspecified methods (6), while game-based interventions were categorized as gamification (4), serious games (8), and digital games (8). Reported outcomes indicated improvements in motivation, therapy adherence, and various metrics of rehabilitation performance. Nonetheless, key challenges persist, including small sample sizes, methodological heterogeneity, data privacy concerns, and the lack of large-scale clinical trials. Barriers related to cost, accessibility, and ethics also hinder broader implementation. Future research should prioritize multimodal data integration, AI-driven remote rehabilitation platforms, and wearable technologies to enhance scalability and clinical impact.