Research Project
6858105
DESCRIPTION (provided by applicant): This exploratory proposal aims to develop and investigate a novel means of determining the risk of fracture of skeletal systems. The problem of increased risk of fracture as a result of bone mass loss due to aging or disease is a major clinical concern, and has been the focus of a substantial research effort. Although some correlative relationships have been established between fracture risk and various imaging and biochemical data, these methods remain nonspecific and have low sensitivity for predicting fracture risk. Clearly, an accurate means of predicting risk of fracture for an individual subject would have tremendous clinical benefit. We propose to combine several analysis techniques that have not been used together before to produce a rigorously verified and validated, physics based approach to quantify the probability or risk of bone fracture. Namely, probabilistic finite element methods will provide a framework for accounting for uncertainty and randomness inherently involved in the mechanics of skeletal structures, and a recently developed constitutive model will describe the time-dependent, permanent, and stiffness degradation behavior of bone. Models of vertebral trabecular bone specimens and whole vertebrae will be produced using geometry and material property descriptions based on high-resolution micro-CT imaging data, allowing quantitative evaluation of predictive accuracy for specimen-level and whole-bone level failure, and for determining the probability of structural failure with respect to loading history. Although, initially applied to models to vertebral fractures, these methods have tremendous potential for providing individualized interpretive models to predict fracture risk for the discrete patient with a higher degree of accuracy than large population statistical correlations can provide. Further, this exploratory study will provide important information to guide diagnostic imaging and constitutive model refinements, in order to fully utilize the inherent potential of these probabilistic imaging-based methods.
