AI-Guided Virtual Physiotherapy vs. Traditional Therapy for Post-Stroke Rehabilitation: A Randomized Controlled Trial

Abstract

Background: Stroke remains a leading cause of long-term disability, with motor and functional impairments requiring intensive rehabilitation. Conventional physiotherapy is effective but limited by accessibility and adherence. The integration of artificial intelligence (AI) into remote rehabilitation platforms offers an innovative approach to optimize recovery while addressing barriers of traditional therapy. Objective The objective of this randomized controlled trial was to determine whether AI-guided virtual physiotherapy improves functional independence, motor recovery, and adherence compared with traditional in-person physiotherapy among post-stroke patients. Methods A total of 100 patients with ischemic or hemorrhagic stroke were randomized into AI-guided virtual physiotherapy (n=50) and traditional therapy (n=50) groups. Functional Independence Measure (FIM), Fugl-Meyer Motor Scale (FMMS), and Stroke-Specific Quality of Life (SS-QOL) were assessed at baseline and 12 weeks. Adherence rates were also recorded. Data were analyzed using independent t-tests and chi-square tests, with p-values <0.05 considered statistically significant. Results Patients in the AI-guided group showed significantly greater improvements in functional independence (mean improvement in FIM: 28.6 vs. 21.4, p=0.01) and motor recovery (mean improvement in FMMS: 19.8 vs. 15.2, p=0.02). Quality of life improved more in the AI-guided group (24% vs. 16%, p=0.03). Adherence was higher in the AI group, with 88% achieving ≥85% adherence compared with 72% in the traditional group (p=0.04). Conclusion AI-guided virtual physiotherapy demonstrated superior functional, motor, and quality-of-life outcomes with higher adherence compared to traditional therapy, highlighting its potential as an effective and scalable post-stroke rehabilitation strategy.

Ähnliche Arbeiten