Interpretable Machine Learning on Simulation‐Derived Biomechanical Features for Hamstrings–Quadriceps Imbalance Detection in Running

Hamstrings–quadriceps (H–Q) imbalance represents a biomechanical marker of knee instability and injury risk in running. This in silico study introduces a simulation-derived machine-learning framework designed to estimate H–Q imbalance using biomechanical features conceptually mappable to inertial-sensor domains. A reduced musculoskeletal framework emulating flexor–extensor balance, limb symmetry, and co-contraction patterns generated 573 synthetic running trials for 160 virtual subjects across three speeds. These interpretable features trained a calibrated gradient-boosting classifier evaluated via ROC-AUC, PR-AUC, balanced accuracy, F1, and Brier score. Across all conditions, the model achieved ROC-AUC 0.933 (95% CI 0.908–0.958), balanced accuracy 0.943, PR-AUC 0.918, F1 0.940, and Brier 0.056, outperforming a calibrated logistic baseline. Dynamic H:Q ratio and knee-moment symmetry were the dominant predictors, while co-contraction contributed complementary nuance. These results indicate that simulation-derived digital frameworks can reproduce IMU-relevant biomechanical variability, enabling interpretable machine learning for objective assessment of muscular balance in sports medicine.