Assessment of Hip Fracture Risk With an Image-based Three-Level Biomechancial Model

Abstract

Accurate assessment of hip fracture risk is necessary for taking appropriate preventive and protective measurements. Conventional population-based statistical models are not accurate in predicting fracture risk in individual subjects. Single-level biomechanical models are not able to accommodate all biomechanical variables that take effects at different musculoskeletal levels. We developed an image-based three-level biomechanical model to predict hip fracture risk caused by sideways fall from standing height. A small-scale clinical study was conducted to examine the ability of the three-level model to discriminate clinical hip fracture cases from matched controls. The obtained results show that the model has superior performance than the existing clinical tools.