AI ANALYSIS REPORT

Utilization of Machine Learning to Identify Lower Extremity Biomechanical Predictors of Rupture in a Validated Cadaveric Model of ACL Injury

Anterior cruciate ligament (ACL) rupture is a critical concern in sports medicine, often entailing prolonged recovery. This study evaluates eight distinct machine learning (ML) models—including Support Vector Machines, Decision Tree Classifiers, Random Forest, Stochastic Gradient Descent, Logistic Regression, Gradient Boosting, Ridge Regression, and Linear Discriminant Analysis—for predicting ACL injuries. Models were trained and tested on four datasets (ARBD, BARBD, ARW, BARW) with features ranging from 13 (wearable-compatible) to 53 (biomechanical and demographic). We utilized both a three-class (pre-rupture, trial prior to rupture, rupture) and a binary (pre-rupture vs. elevated risk) classification schema. Our findings highlight early-phase force metrics (e.g., 33ms_Fx and 33ms_Fz) and initial-contact forces (e.g., IC_Fx, IC_Fz) as consistent significant predictors across models. Accuracies ranged from 79% to 87% for three-class classification, improving significantly to 92% to 95% for binary classification. These results underscore the clinical relevance of early dynamic measurements and demonstrate the robustness of our ML approach for ACL injury prediction.