REVOLUTIONIZING REHABILITATION: THE ROLE OF ARTIFICIAL INTELLIGENCE IN MODERN PHYSIOTHERAPY

Abstract

Background: Incorporating Artificial Intelligence in rehabilitation and physiotherapy is deemed to be one of the areas where AI may assist in robotization. Undoubtedly, AI has the potential to autonomously personalize treatment plans, enhance rehabilitation primary methods, and improve patient outcomes. Notwithstanding, the professional groups have varying perceptions of the benefits of AI to the physiotherapy practice. For better utilization of AI in rehabilitation practice, understanding the reasons behind these differences is vital. Objectives: This research aims to analyze the level of awareness and adoption of AI in physiotherapy which includes ascertaining the extent to which its implementation is perceived to be beneficial or challenging and its statistical correlate to familiarity with AI, professional background, and readiness to accept AI-based rehabilitation aides. Methods: The quantitative method was applied by gathering descriptive data across a segment to take the cross-sectional survey from a sample of 273 physiotherapists, medical doctors, patients, and artificial intelligence researchers. Questions were prepared to retrieve information relating to familiarity with AI and its perception, adoption, and barriers. The gathered data was processed through descriptive and inferential statistics using the Chi-square test, T-test ANOVA, and reliability analysis (Cronbach’s Alpha). Shapiro-Wilk and D’Agostino K2 Normality Tests were utilized to analyze the distribution of responses, Levene’s test measured variation among professional group responses. Results: The results showed that the adoption of AI in physiotherapy is not linked to the user’s professional background (Chi-square p > 0.05) meaning, understanding the usages resides on personal exposure, not the individual’s profession. In contrast, Levene’s test (p < 0.05) found differences across professional groups’ AI knowledge, showing that some have more understanding of its use than others. The normality tests revealed the presence of a bias in the distribution, with regards to AI, where respondents were grouped into optimists and skeptics. Furthermore, Cronbach’s Alpha (-0.18) demonstrated weak reliability on the scale responses, indicating that participants were polled with AI questions inconsistently. Primary reasons for which AI is not adopted include lack of knowledge, adequate training, and monetary and ethical issues. Conclusion: This study shows there are notable gaps in the adoption of AI in physiotherapy which highlights the need for education and training with integrated standard strategies. AI has the potential to transform rehabilitation programs, as such there is a need to increase awareness, access, and policy for its usage in physiotherapy. Further research should be directed toward AI’s longitudinal effects on patient recovery, costs for therapy, and overall efficiency of the healthcare system through experimental trials. Tackling these issues will increase the effectiveness and reliability of AI-driven solutions, thus enabling more widespread adoption in clinical practice.