KNOWLEDGE AND ADOPTION OF AI-ENHANCED REHABILITATION TECHNOLOGIES AMONG PHYSICAL THERAPISTS: A CROSS-SECTIONAL STUDY

Abstract

Background: Artificial intelligence (AI) has begun to reshape rehabilitation practices by offering advanced tools such as motion sensors, virtual feedback systems, and intelligent assessment technologies. Despite growing innovation, the clinical integration of AI in physical therapy remains unclear, particularly in resource-variable settings such as Lahore, Pakistan.

Objective: To investigate the familiarity, perceived usefulness, and implementation of AI-enhanced rehabilitation tools among physical therapists in clinical settings.

Methods: A cross-sectional survey was conducted over eight months across hospitals, private clinics, and academic-affiliated institutions in Lahore. A total of 350 licensed physical therapists were recruited using stratified sampling. A validated self-administered questionnaire assessed participants’ knowledge, frequency of use, perceived barriers, and usefulness of AI tools. Descriptive statistics, t-tests, ANOVA, and Pearson’s correlation were used for data analysis.

Results: The mean age of participants was 32.6 ± 6.8 years, with an average of 8.1 ± 5.3 years of clinical experience. Familiarity with AI tools was highest for motion sensors (31.2% high familiarity), while AI-based assessment tools had the lowest familiarity (19.5%). Clinical use was limited, with only 13.8–24.1% of respondents reporting frequent use. Major barriers included lack of training (72.4%) and equipment cost (65.3%). Motion sensors were rated highest in usefulness (mean score 3.9/5), correlating with higher usage and familiarity.

Conclusion: There is moderate awareness but low clinical implementation of AI in physical therapy. Addressing barriers through training and institutional support is crucial for improving adoption and enhancing patient outcomes.