Predictive Modeling of Joint Angles Using Machine Learning: A Comparative Study of Random Forest and Support Vector Machine

In biomechanics, accurate prediction of joint angles under varying conditions is essential for the optimization of rehabilitation protocols, orthopedic design, and athletic performance. Joint movements are inherently complex, which are influenced by various external factors and the specific joint being observed. Previous studies have employed machine learning techniques, such as Random Forest (RF) and Support Vector Machines (SVMs), to accurately predict joint angles under conditions of simple movements and gait analysis, which used data from wearable sensors such as Inertial Measurement Units (IMUs) and Surface ElectroMyoGraphy (SEMG). While these models have been effective in predicting joint angles in real-time applications and movement analysis, no prior research has focused on the prediction of joint angles under varying brace conditions, such as unbraced, knee brace, and ankle brace. This limits the understanding of how external supports affect joint movement dynamics. To address this limitation, this study investigates the use of RF and SVM in predicting joint angles under three specific conditions: unbraced, knee brace, and ankle brace. Feature engineering techniques are applied to transform the sequential data; the models are evaluated based on accuracy, robustness, and computational efficiency. The results will offer insights on the impact of braces on joint movement, which contribute to the literature on machine learning in biomechanics. This research has practical implications for rehabilitation, the design of wearable orthopedic devices, athletic training, which advances the predictive modeling of joint biomechanics.

Mohammad Samara holds a Bachelor's and Master's degree in Industrial Engineering and is currently pursuing a PhD at Binghamton University. He has professional experience as a Quality Assurance and Production Engineer across diverse industries. His research focuses on applying Lean Six Sigma methodologies to optimize healthcare systems, and he has worked extensively on enhancing service delivery systems. Additionally, Mohammad specializes in data analysis, utilizing statistical and machine learning techniques to drive continuous improvement in healthcare.