Advancing public health through artificial intelligence in physiotherapy: a bibliometric analysis

Abstract

This scientometric study investigates global trends and applications of artificial intelligence (AI) in physiotherapy, including rehabilitation, movement analysis, and telerehabilitation, between 2006 and 2024. The increasing applications of AI in physiotherapy may have the potential to enhance outcomes, increase operational efficiency, and bring innovative solutions. However, the rapid evolution of AI technologies and applications in physiotherapy necessitates critical examination of patterns of research and geographic differences. The articles were published and indexed in Web of Science and Scopus, and data collection was done in July 2024. A thorough search query was employed: TITLE-ABS-KEY ("Artificial Intelligence" OR "AI") AND ("Physiotherapy" OR "Physio" OR "PT"), which yielded 377 records. Excluding 35 duplicate records and 2 non-eligible studies screened, 340 studies were included. The inclusion criteria were peer-reviewed journal articles, conference articles, and review articles in the English language with a minimum of five citations. Exclusion criteria ruled out non-peer-reviewed articles, editorials, and non-physiotherapy-related studies. The analysis identifies a uniform rate of growth in research productivity with a steep rate of growth between 2019 and 2022 with advances in artificial intelligence technologies, such as machine learning, wearable technology, and robot rehabilitation. The major research institutions were mapped in North America, Europe, and Asia; however, significant geographic disparities exist. Bibliometric indicators, such as H-index, collaboration networks, and co-authorship analysis, were used to quantify the productivity of authors, journals, and institutions. AI has shown its potential to transform physiotherapy, in particular, to maximize rehabilitation and treatment outcomes. The review does note, however, that future studies should consider ethical factors, such as data privacy, algorithmic bias, and explainability. The findings suggest the significance of exploring the long-term clinical impact of AI and interdisciplinary collaboration to address regional inequalities. Research needs to be carried out in underexplored areas, such as pediatric rehabilitation and AI-based decision-making systems, to ascertain even and effective AI integration into physiotherapy practice among populations. • The study presents the first bibliometric analysis of AI in physiotherapy with a 15.9 percent annual growth. • It explores AI applications in physiotherapy, including wearable devices, pose estimation and robotics. • The study reveals geographical gaps as North America, Europe and Asia lead while Africa and South America lag. • It highlights ethical concerns such as data privacy, algorithmic bias and the need for transparent AI systems. • The work identifies trends in deep learning and neural networks and calls for stronger global collaboration.

Ähnliche Arbeiten