Awareness of Machine Learning for Shoulder Exercises and Assessment in Students and Physiotherapists – A Cross-Sectional Study

Abstract

Background: Machine learning (ML), as a core domain of artificial intelligence, has recently emerged as a promising tool for the analysis, diagnosis, and rehabilitation of shoulder conditions in physiotherapy. Despite its global growth, its integration remains limited in low- and middle-income countries, including Pakistan. Understanding local professionals’ perceptions and readiness is essential for guiding future adoption. Objective: To assess the perceptions, awareness, and readiness of physiotherapists in Lahore, Pakistan, toward using ML-based tools for shoulder exercise evaluation and rehabilitation. Methods: A descriptive cross-sectional study was conducted from January to June 2025 among 150 physiotherapists recruited through convenience sampling from universities and clinical institutes in Lahore. A pre-validated, five-point Likert scale questionnaire assessed participants’ awareness, perceptions, and readiness to integrate ML into shoulder rehabilitation practice. Data were analyzed using SPSS, employing descriptive statistics, chi-square tests, and correlation analyses. Statistical significance was set at p < 0.05. Results: Most respondents were younger than 25 years (68%), with an equal distribution of males and females. Overall, participants demonstrated moderate awareness of ML applications in physiotherapy, with positive perceptions toward its potential role in improving assessment accuracy and treatment outcomes. Readiness to adopt ML tools showed significant associations with age, professional exposure, and previous familiarity with digital technologies (p < 0.05). Conclusion: Physiotherapists in Lahore show favorable perceptions and emerging readiness to integrate ML into shoulder rehabilitation, highlighting the need for targeted training, institutional support, and accessible technological resources.