Exploring rehabilitation therapists’ knowledge and perspectives on the use of artificial intelligence and machine learning for persons post-stroke

Abstract

Background The integration of artificial intelligence (AI) and machine learning (ML) within the clinical work of rehabilitation therapists in post stroke rehabilitation has the potential to transform their work performance, habits, and routines. The impact could potentially enhance integration of assessment/data analysis, personalizing therapies, documentation, and promoting home program adherence. Objective This study assesses occupational and physical therapists’ current perspectives on AI/ML in post-stroke clinical rehabilitation and examines perceived barriers to potential application of AI/ML in their clinical work. Methods This mixed methods study includes quantitative survey data collected pre/post an AI/ML educational webinar, as well as qualitative focus group analysis that includes therapist attitudes and experiences with AI/ML, practical applications to using AI/ML, as well as the difficulties associated. The recruited therapists include occupational and physical therapy practitioners from the United States and Asia. Individuals included in the study had to have a caseload of 30% persons with stroke. Results Four key themes emerged from qualitative analysis: 1) current clinical work practice and identified gaps, 2) current therapists’ perceptions of AI/ML, 3) perceived future of AI/ML in clinical work practices and 4) steps needed for implementation. Survey data reporting frequencies of technology use and AI/ML perceptions are integrated into qualitative findings. Conclusions Therapists can articulate potential future uses for AI/ML technology in post-stroke rehabilitation, however, remain uncertain about when, how, and at what cost utilization will be supported in clinical work environments. Therapists desire education and support to utilize the evolving technologies in practice.