| Applications of Artificial Intelligence in Physical Therapy: A Summary and Directions for Future Research |
| Paper ID : 1011-ISCSR3 (R1) |
| Authors |
|
Ahmed Ibrahim Rashad * Faculty of Physical Therapy, Cairo University |
| Abstract |
| Background: In the past few decades, Artificial Intelligence (AI) has garnered much attention in healthcare, including the field of physical therapy. Previous reviews focused on summarizing the applications in one physical therapy specialty at a time. Up-to-date reviews that encapsulate the applications of AI in multiple departments of physical therapy remain scarce. Objective: This article aims to outline the applications of artificial intelligence in different specialties of physical therapy and to present implications for future research. A secondary objective of this article is to summarize the attitudes and opinions of physical therapists toward artificial intelligence. Methods: A literature search was performed in PubMed and Cochrane Library, using terms synonymous with “Artificial Intelligence” and “Physical Therapy” for studies published up to March 2025. With no restrictions to study design, studies that involved the application of AI in any physical therapy department were selected. Data were extracted using specialized sheets that included columns for the study design, aims, and main findings of the included studies. Results: The plausible applications can be broadly classified into assessment, prediction, and treatment. Assessment encompassed diagnosis of diseases and the measurement of parameters. Prediction involved the prediction of disease progression and treatment outcomes. As for treatment, AI played a pivotal role in telerehabilitation. AI proved to be a plausible and novel tool for adults diagnosed with orthopedic conditions, neurological conditions, and cardiac disorders. AI was also used in pediatrics. The attitudes of physical therapists toward artificial intelligence were mostly optimistic. However, concerns about data privacy problems and the loss of human interaction remain among therapists. Conclusion: Current evidence demonstrates the plausibility of utilizing artificial intelligence in physical therapy albeit more replication is needed. Attention should also be paid to investigating the concerns of physical therapists about data privacy and human interaction. |
| Keywords |
| "Artificial Intelligence", "Deep Learning", "Machine Learning", "Physical Therapy", "Rehabilitation" |
| Status: Abstract Accepted |