Effectiveness of AI-assisted rehabilitation for musculoskeletal disorders: a network meta-analysis of pain, range of motion, and functional outcomes

作者：Zixuan Luo, Yang Wang, Tongyan Zhang, Jiale Wang · 发表于：Frontiers in Bioengineering and Biotechnology · 年份：2025 · DOI：10.3389/fbioe.2025.1660524 · 被引用次数：11 · 研究领域：Stroke Rehabilitation and Recovery、Prosthetics and Rehabilitation Robotics、Artificial Intelligence in Healthcare and Education

Objective: This study aims to compare the effectiveness of 13 artificial intelligence (AI)-assisted rehabilitation strategies for individuals with musculoskeletal disorders (MSDs), categorized based on different intervention types, including AI feedback systems, exergaming platforms, telerehabilitation, and robotic solutions. The analysis focuses on improvements in pain relief, functional outcomes, and range of motion (ROM), based on a network meta-analysis (NMA) of randomized controlled trials (RCTs). Methods: A systematic review and NMA were conducted in accordance with PRISMA guidelines. Four databases (PubMed, Embase, Cochrane Library, Web of Science) were searched for RCTs published between January 2000 and April 2025. A total of 33 RCTs involving participants with MSDs were included. Interventions were categorized into 13 AI-assisted rehabilitation strategies. The outcomes were grouped into three domains: pain, functional outcomes, and ROM. Surface under the cumulative ranking curve (SUCRA) values and mean ranks were used to compare the relative effectiveness of each intervention. The Risk of Bias (RoB 2) tool was used to assess the bias risk of the studies, and the Confidence in Network Meta-Analysis (CINeMA) tool was applied to evaluate the credibility of the evidence. Results: For pain relief, Therapeutic Exergaming (SUCRA = 87.6%) and Robotic Exoskeleton (SUCRA = 86.3%) ranked highest. In functional outcomes, Gamified Exergaming (SUCRA = 99.6%) and Hybrid Physical T...