Determine and compare limb specific predictors based on the foot pressure index of right and left foot among young adults in Saudi Arabia using machine learning models.

Foot posture is a key determinant of limb-biomechanics, gait balance as well as functional mobility. Usage of foot pressure index (FPI) is widespread to but the determinants of FPI without accounting to the left right foot differences. Therefore, identifying limb-specific determinants would enhance screening accuracy and help in more targeted interventions. The objective of the study was to determine and compare limb specific predictors based on the foot pressure index of right and left foot among young adults using machine learning models. A total of 581 (290 females and 291 males) 18-25-year-old young adults were assessed on anthropometric measures, joint hyper mobility and functional mobility. Using SAS-Viya, separate entropy-based decision trees, along with logistic regression models were developed and model performance was evaluated using validation-test accuracy and area under the curve (AUC). Decision tree models outperformed logistic regression for both FPI-Left (FPI-L) and FPI-Right (FPIR). FPIR was primarily predicted by joint hyper mobility, followed by stair ascent time, stability as well as BMI with stronger influence of both structural and load related factors. FPIL was predicted by closed-eye balance along BMI, open eye balance, stair ascent, Beighton score, underscoring the role of sensory motor control. The findings of the study reflect that young adults foot posture is complex, limb specific interactions among anthropometric, functional and sensory factors.