Studi Literatur Penggunaan Artificial Intelligence untuk Terapi Gangguan Muskuloskeletal

Abstract

Abstract

Background: The application of artificial intelligence (AI) in physiotherapy and musculoskeletal rehabilitation has developed rapidly due to its potential to improve service accessibility, therapy personalization, patient education, and exercise monitoring. However, the consistency of clinical recommendations and the continued need for professional supervision remain important concerns.

Objective: To review the scientific evidence regarding the use of AI in the management of musculoskeletal disorders, particularly in the context of rehabilitation and clinical decision-support systems.

Methods: This article is a narrative literature review. The literature search was conducted in the Scopus database using the keywords “musculoskeletal disorder” AND “artificial intelligence”. Eligible studies included research articles published between 2020 and 2025 in scientific journals and relevant to the application of AI in musculoskeletal disorders. From an initial 251 publications, studies were screened based on article type, year of publication, title, abstract, study objectives, results, and availability of full-text manuscripts, resulting in 8 articles included in the analysis.

Results: The eight selected articles demonstrated that AI has been applied in various forms, including large language models such as GPT-4, digital care programs, AI-based patient-reported outcome measures, and computer vision systems for movement analysis. Overall, AI showed considerable potential in patient education, exercise recommendation, rehabilitation monitoring, expansion of healthcare service coverage, and pain reduction in several musculoskeletal conditions. GPT-4 was reported to outperform GPT-3.5 in diagnosis and treatment planning; however, its consistency with rehabilitation guidelines has not yet been fully optimized.

Conclusion: Artificial intelligence has the potential to serve as a valuable adjunctive tool in the management of musculoskeletal disorders, particularly for education, monitoring, and rehabilitation personalization. Nevertheless, its implementation still requires clinical validation and professional oversight, ensuring that AI functions as a supportive tool rather than a replacement for physiotherapists.

Keywords: artificial intelligence, physiotherapy, musculoskeletal, rehabilitation, GPT-4.