Application of metamodeling for characterization of bone-remodeling parameters using clinical results
Gabriela Wessling Oening Dicati1; José Eduardo Gubaua1; Bárbara Belisa Soffiati1; Jucélio Tomás Pereira1
1 Universidade Federal do Paraná
doi:10.20906/CPS/CILAMCE2017-1009
Resumo
The bone tissue has the capacity for adaptation and renewal and this process is named bone remodeling. Many mathematical models have been developed with the purpose of describing this process, being applied to several bones, especially to the human femur. These models can be used to simulate different tissue characteristics, such as bone morphology and material anisotropy in the longitudinal direction of loading. In addition, these models allow a visualization of the different levels of porosity, which characterize the trabecular and cortical bone tissues. As a result of these models, there is a distribution of densities that can be compared qualitatively with an X-ray, being dependent of the parameters of the mathematical model used. The major question is that the femoral densities distribution is unique to each individual. Due to the parameters determined of an empirical way, such as the width and central position of the dead zone, the mathematical model does not characterize this individuality of a quantitative way. Thus, the objective of this study is to determine the best parameters of the Stanford bone remodeling model with isotropic behavior, modifying the width and central position of the dead zone, in order to obtain a distribution of densities as similar as possible with a computerized tomography. For the development of this study, concepts of metamodeling and optimization are employed. The construction of the solid model of the femur and the assignment of densities are performed through the image files of a computed tomography. Numerical simulations of bone remodeling are performed using the Finite Element Method via Abaqus software. The results of the article present promising results, allowing the characterization of the femur and a quantitative comparison of the distribution of densities with a computed tomography.
Palavras-chave: bone remodeling; Finite Element Method ; Characteristic Parameters; Metamodeling; Computed Tomography