Researchers now report an integrated experimental–computational framework for generating continuum material maps as a function of density, built from tissue-scale characterization of the canine femur. The approach combines high-depth spherical nanoindentation with Genetic Algorithm (GA)-based inverse optimization to extract mechanically consistent elasto-plastic properties of cortical and trabecular bone. The optimization works directly from load–displacement responses that are influenced by post-yield contact uncertainties, allowing the tissue-level properties to be recovered despite the complications that arise when indentation pushes material past its elastic limit.

With tissue-scale properties in hand, the team assigned trabecular values to Representative Volume Elements (RVEs) reconstructed from micro-Quantitative Computed Tomography (μQCT) data. These RVEs were then homogenized to capture architecture-dependent effective behavior, meaning the framework accounts for how the porous microstructure of trabecular bone alters its overall mechanical response without having to resolve that microarchitecture explicitly in a whole-bone model.

In the final step, the optimized cortical properties and the homogenized trabecular properties were correlated with density to establish compartment-specific material mapping relationships. These relationships are intended for use in whole-bone FEA simulations, where spatially varying material inputs are assigned across the femur based on density distributions obtained from imaging.

The optimized responses showed good agreement with the experimental indentation measurements, supporting the capability of the proposed framework to generate mechanically informed material definitions for patient-specific bone FEA. According to the authors, the framework provides a pathway for translating localized experimental measurements into heterogeneous elasto-plastic material inputs for whole-bone FEA without explicitly resolving trabecular microarchitecture. That translation step is significant for veterinary and translational applications, since it links tissue-level elasto-plastic characterization—capturing behavior beyond the yield point—to continuum models that can be run at the scale of the entire bone. The work was conducted by Behzad Amirzade, Tareq Zobaer, Janis Lapsley, Do-Gyoon Kim, Laura E. Selmic, and Ali Nassiri, and addresses a persistent gap between what can be measured at the tissue scale and what simplified material models typically assume in patient-specific simulations of both human and veterinary bone mechanics.

Subject of Research: Elasto-plastic characterization and density-based material mapping of canine bone for finite element analysis

Article Title: An integrated experimental–computational framework for elasto-plastic characterization and spatial mapping of canine bone properties

Article References: Amirzade, B., Zobaer, T., Lapsley, J., Kim, D.-G., Selmic, L. E., & Nassiri, A. (2026). An integrated experimental–computational framework for elasto-plastic characterization and spatial mapping of canine bone properties. PLOS One, 21(10), e0361283. https://doi.org/10.1371/journal.pone.0361283

Image Credits: AI Generated

DOI: 10.1371/journal.pone.0361283

Keywords: bone mechanics, finite element analysis, nanoindentation, genetic algorithm, μQCT, trabecular bone, cortical bone, canine femur, elasto-plasticity, material mapping, integrated, experimental