Personalized loading conditions for homogenized finite element analysis of the distal sections of the radius

• DOI: 10.1007/s10237-022-01656-4
• Publication Date: 2022-12-07T09:08:51Z
• Authors: Denis Schenk; Philippe Zysset
• Keywords: Homogenization

The microstructure of trabecular bone is known to adapt its morphology in response mechanical loads for achieving a biomechanical homeostasis. Based on this form-function relationship, previous investigators either simulated the remodeling predict resulting density and architecture specific loading or retraced physiological conditions from local architecture. latter inverse approach includes quantifying using computed tomography calculating relative importance selected load cases by minimizing fluctuation tissue level metric. Along concept, present study aims at identifying an optimal, personalized, multiaxial case distal section human radius vivo HR-pQCT-based isotropic, homogenized finite element (hFE) analysis. dataset consisted HR-pQCT reconstructions 20 mm most 21 fresh-frozen radii. We six different unit canonical (FX palmar-dorsal force, FY ulnar-radial FZ distal-proximal MX moment about palmar-dorsal, MY ulnar-radial, MZ distal-proximal) simplified efficient hFE method based single isotropic phase. Once we used homogeneous mean (shape model) once original heterogeneous distribution + model). Using analytical formulation, minimized deviation strain tensors ε(x) hydrostatic compressive reference ε0, 6 degrees freedom (DOF) optimal (OPT) all individual 1 DOF (FX, FY, FZ, MX, MY, MZ). All seven were then extended nonlinear regime scaled displacements linear as boundary (MAX). compared models their objective function (OF) values, stored energies ultimate strength torsor norm. Both shape linear-optimized OPT dominated positive force z-direction (FZ). Transversal DOFs close zero depending model type. inclusion increased influence changed direction while was small both models. had 12-15% lower values than case, model. Stored optimum consistently 142-178% higher case. Differences maximum norm ‖t‖ heterogeneous, but OPT_MAX FZ_MAX. presented proof concept optimization procedure estimate patient-specific methods. In contrast similar models, included metric that favors compression. analysis joint surfaces radius, estimated directions are plausible. For our dataset, standard axial compression conditions, usually FE today. But even model, optimized achieves more can absorb twice energy uniaxial calculated with 6-DOF (OPT_MAX) 1-DOF (FZ_MAX). Defining may decrease angulation errors during CT measurements improve repeatability well reproducibility stiffness These results encourage extension anisotropic application data sets test hypothesis reduced measurement.