μFE Models Research Articles

Computational simulation of trabecular adaptation progress in human proximal femur during growth

There are a large number of clinical and experimental studies that analyzed trabecular architecture as a result of bone adaptation. However, only a limited amount of quantitative data is currently available on the progress of trabecular adaptation during growth. In this paper, we proposed a two-step numerical simulation method that predicts trabecular adaptation progress during growth using a recently developed topology optimization algorithm, design space optimization (DSO), under the hypothesis that the mechanisms of DSO are functionally equivalent to those of bone adaptation. We applied the proposed scheme to trabecular adaptation simulation in human proximal femur. For the simulation, the full trabecular architecture in human proximal femur was represented by a two-dimensional μFE model with 50 μm resolution. In Step 1, we determined a reference value that regulates trabecular adaptation in human proximal femur. In Step 2, we simulated trabecular adaptation in human proximal femur during growth with the reference value derived in Step 1. We analyzed the architectural and mechanical properties of trabecular patterns through iterations. From the comparison with experimental data in the literature, we showed that in the early growth stage trabecular adaptation was achieved mainly by increasing bone volume fraction (or trabecular thickness), while in the later stage of the development the trabecular architecture gained higher structural efficiency by increasing structural anisotropy with a relatively low level of bone volume fraction (or trabecular thickness). We demonstrated that the proposed numerical framework predicted the growing progress of trabecular bone that has a close correlation with experimental data.

A numerical model of heterogeneous surface strains in polymer scaffolds

In vitro bone tissue growth inside porous scaffolds can be enhanced by macroscopic cyclic compression of the construct, but the heterogeneous strain generated inside the construct must be investigated to determine appropriate levels of compression. For this purpose a linear micro-finite element (μFE) technique based on micro-computed tomography (μCT) was verified for the calculation of local displacements inside polymer scaffolds, from which local strains may be estimated. Local displacements in the axial direction at the surface of microstructures inside the scaffold in 60 locations were calculated with the μFE model, based on compression simulation of a μCT reconstruction of the scaffold. These displacements were compared with accurately measured displacements in the axial direction in the same polymer scaffold at the same 60 locations, using a micro-compression chamber and μCT reconstructions of the scaffold under two fixed levels of compression (5% and 0%). The correlation between the calculated and the measured displacements, after correction for the dependence of the axial displacement on the axial position, was r=0.786 ( r 2=0.617). From this we conclude that the linear μFE model is suitable to estimate local surface strains inside polymer scaffolds for tissue engineering applications. This technique can not only be used to determine appropriate parameters such as the level of macroscopic compression in experimental design, but also to investigate the cellular response to local surface strains generated inside three-dimensional scaffolds.

A scalable multi‐level preconditioner for matrix‐free µ‐finite element analysis of human bone structures

AbstractThe recent advances in microarchitectural bone imaging disclose the possibility to assess both the apparent density and the trabecular microstructure of intact bones in a single measurement. Coupling these imaging possibilities with microstructural finite element (µFE) analysis offers a powerful tool to improve bone stiffness and strength assessment for individual fracture risk prediction.Many elements are needed to accurately represent the intricate microarchitectural structure of bone; hence, the resulting µFE models possess a very large number of degrees of freedom. In order to be solved quickly and reliably on state‐of‐the‐art parallel computers, the µFE analyses require advanced solution techniques. In this paper, we investigate the solution of the resulting systems of linear equations by the conjugate gradient algorithm, preconditioned by aggregation‐based multigrid methods. We introduce a variant of the preconditioner that does not need assembling the system matrix but uses element‐by‐element techniques to build the multilevel hierarchy. The preconditioner exploits the voxel approach that is common in bone structure analysis, and it has modest memory requirements, at the same time robust and scalable.Using the proposed methods, we have solved in 12min a model of trabecular bone composed of 247 734 272 elements, yielding a matrix with 1 178 736 360 rows, using 1024 CRAY XT3 processors. The ability to solve, for the first time, large biomedical problems with over 1 billion degrees of freedom on a routine basis will help us improve our understanding of the influence of densitometric, morphological, and loading factors in the etiology of osteoporotic fractures such as commonly experienced at the hip, spine, and wrist. Copyright © 2007 John Wiley & Sons, Ltd.

Minimal size of structural μFE models of trabecular bone to predict the apparent stiffness

Minimal size of structural μFE models of trabecular bone to predict the apparent stiffness

Estimation of distal radius failure load with micro-finite element analysis models based on three-dimensional peripheral quantitative computed tomography images

There is increasing evidence that, in addition to bone mass, bone microarchitecture and its mechanical load distribution are important factors for the determination of bone strength. Recently, it has been shown that new high-resolution imaging techniques in combination with new modeling algorithms based on the finite element (FE) method can account for these additional factors. Such models thus could provide more relevant information for the estimation of bone failure load. The purpose of the present study was to determine whether results of whole-bone micro-FE (μFE) analyses with models based on three-dimensional peripheral quantitative computer tomography (3D-pQCT) images (isotropic voxel resolution of 165 μm) could predict the failure load of the human radius more accurately than results with dual-energy X-ray absorptiometry (DXA) or bone morphology measurements. For this purpose, μFE models were created using 54 embalmed cadaver arms. It was assumed that bone failure would be initiated if a certain percentage of the bone tissue (varied from 1% to 7%) would be strained beyond the tissue yield strain. The external force that produced this tissue strain was calculated from the FE analyses. These predictions were correlated with results of real compression testing on the same cadaver arms. The results of these compression tests were also correlated with results of DXA and structural measurements of these arms. The compression tests produced Colles-type fractures in the distal 4 cm of the radius. The predicted failure loads calculated from the FE analysis agreed well with those measured in the experiments ( R 2 = 0.75 p < 0.001). Lower correlations were found with bone mass ( R 2 = 0.48, p < 0.001) and bone structural parameters ( R 2 = 0.57 p < 0.001). We conclude that application of the techniques investigated here can lead to a better prediction of the bone failure load for bone in vivo than is possible from DXA measurements, structural parameters, or a combination thereof.

High-resolution three-dimensional-pQCT images can be an adequate basis for in-vivo microFE analysis of bone.

Micro-finite element (microFE) models based on high-resolution images have enabled the calculation of elastic properties of trabecular bone in vitro. Recently, techniques have been developed to image trabecular bone structure in vivo, albeit at a lesser resolution. The present work studies the usefulness of such in-vivo images for microFE analyses, by comparing their microFE results to those of models based on high-resolution micro-CT (microCT) images. Fifteen specimens obtained from human femoral heads were imaged first with a 3D-pQCT scanner at 165 microns resolution and a second time with a microCT scanner at 56 microns resolution. A third set of images with a resolution of 165 microns was created by downscaling the microCT measurements. The microFE models were created directly from these images. Orthotropic elastic properties and the average tissue von Mises stress of the specimens were calculated from six FE-analyses per specimen. The results of the 165 microns models were compared to those of the 56 microns model, which was taken as the reference model. The results calculated from the pQCT-based models, correlated excellent with those calculated from the reference model for both moduli (R2 > 0.95) and for the average tissue von Mises stress (R2 > 0.83). Results calculated from the downscaled micro-CT models correlated even better with those of the reference models (R2 > 0.99 for the moduli and R2 > 0.96 for the average von Mises stress). In the case of the 3D-pQCT based models, however, the slopes of the regression lines were less than one and had to be corrected. The prediction of the Poisson's ratios was less accurate (R2 > 0.45 and R2 > 0.67) for the models based on 3D-pQCT and downscaled microCT images respectively). The fact that the results from the downscaled and original microCT images were nearly identical indicates that the need for a correction in the case of the 3D-pQCT measurements was not due to the voxel size of the images but due to a higher noise level and a lower contrast in these images, in combination with the application of a filtering procedure at 165 micron images. In summary: the results of microFE models based on in-vivo images of the 3D-pQCT can closely resemble those obtained from microFE models based on higher resolution microCT system.