Modelling woven bone stiffness: a computational image-based approach applied to osteoporotic distraction callus
Please login to view abstract download link
Woven bone is a disorganized tissue of clinical relevance in bone regeneration Its stiffness evolution can be characterized using numerical models based on computed tomography (CT) images [1]. However, current mesh‑generation strategies often rely on complex commercial software, and no consensus exists on how to assign mechanical properties from CT grey levels or mineral density. This work introduces a novel CT‑based finite element methodology that assigns to CT-voxel sized elements mechanical properties from experimental data to predict callus stiffness in osteoporotic patients. Ex vivo samples from different postoperative time points in a previous osteoporotic bone transport study [2, 3] were analysed. Callus morphology was segmented and meshed using a custom Matlab (MathWorks, USA) code. Mechanical properties were assigned based on bone mineral density using a correlation between mineral density and woven bone elastic modulus derived from micro‑CT measurements. Callus mechanical behaviour under in vivo loading conditions was simulated in Abaqus (Dassault Systèmes, France) to estimate the apparent stiffness at each time point. The results show that the proposed methodology reproduces complex morphological geometries with low computational cost and without mesh refinement requirements. Predicted stiffness values agree with in vivo measurements from the same specimens. This approach was further applied to compare osteoporotic and healthy distraction callus [1], preliminarily revealing a reduction in apparent stiffness in osteoporotic animals. Authors would like to acknowledge the funding (PID2023-148828OB-I00 funded by MICIU/AEI/10.13039/501100011033 and by the ERDF/EU; Andalusian Government via Grant DGP_PIDI_2024_01054).
