AI‐enhanced physiotherapy : revolutionizing movement assessment with enhanced functional screenings

Abstract

Artificial Intelligence (AI) has made major strides in a number of sports and healthcare technology fields, providing game‐changing solutions for injury prevention and athlete performance. The Functional Movement Screen (FMS), a popular instrument for evaluating movement patterns and determining possible injury risks, is one field that stands to gain from such advancements. However, the typical FMS depends on physiotherapists’ subjective assessments, which can be laborious, inconsistent, and difficult to implement consistently in real‐time applications. Because of these drawbacks, an automated, impartial, and effective substitute is required to improve the FMS process’s precision and usefulness. By automating the FMS scoring process through the integration of AI‐powered computer vision and machine learning techniques, this thesis tackles the aforementioned difficulties. The suggested system assesses movement patterns, extracts biomechanical properties, and gives immediate feedback on postural misalignments using live video feeds. In contrast to conventional techniques, the system uses ensemble models, such as K‐Nearest Neighbors, Random Forest, and Support Vector Machines, to accurately score workouts and identify movement patterns. This improves efficiency and accuracy. By reducing the shortcomings of individual classifiers, the combination of these models improves the dependability of the findings and guarantees strong predictions. One of the research’s main accomplishments is an automated system that can outperform current methods in terms of consistency and scalability, earning accuracy ratings of over 85% across the majority of FMS exercises. When it came to complex and dynamic exercises like the Inline Lunge and Rotary Stability tests, the ensemble technique showed exceptional strength. Additionally, an evaluation of computational efficiency showed that the suggested approach balances real‐time processing capabilities with high performance. In addition to advancing AI applications in sports and physiotherapy, this study establishes the foundation for further advancements in automated injury prevention technology.