Exploring Advanced Machine Learning Techniques for Gait Analysis in Genu Valgum

This paper offers an in-depth review of recent developments in Gait analysis, highlighting the application of machine learning (ML) techniques for identifying and managing Genu Valgum. The integration of motion capture systems with machine learning algorithms, especially through Python libraries for landmark detection, has greatly improved the detection and forecasting of abnormal Gait patterns. Our proposed model achieved an accuracy of 78.95%, surpassing traditional approaches. Real-time Gait analysis using MediaPipe demonstrated improved consistency in Q-angle measurements, enhancing precision and reliability in diagnosing Genu Valgum. These developments assist clinicians in diagnosing Genu Valgum and creating customized treatment plans. However, challenges such as model interpretability, limited dataset diversity, and variability in real-world applications remain. This paper discusses potential future research directions, emphasizing the need for more advanced algorithms, larger and more diverse datasets, and improved methods for real-time Gait analysis using software-based landmark detection.