Improved anisotropic elastic properties evaluation of cancellous bone using novel experimental and image-based fabric tensor analysis

Accurate evaluation of the mechanical properties of cancellous bone is critical for various clinical diagnoses and orthopedic treatments. In this study, experimental, computational, and analytical investigations are conducted to assess its anisotropic elastic properties. Firstly, a novel elastic modulus measurement method based on the digital image correlation (DIC) technique is proposed to enable reliable determination of the elastic modulus along different directions. Finite element simulations are conducted with both high-resolution micro-CT (micro-CT/FEM) and low-resolution clinical CT (CT/FEM) images-based geometries to explore the effects on elastic modulus predictions. Meanwhile, a fabric tensor-based analytical framework is proposed, employing mean intercept length (MIL) for micro-CT and gray-level structure tensor (GST) method for clinical CT. The results demonstrate that micro-CT/FEM predictions closely match experimental measurements and effectively capture mechanical anisotropy. In contrast, CT/FEM fails to predict transverse and shear moduli accurately. Both MIL- and GST-based models can capture the anisotropic elastic properties reasonably well as micro-CT/FEM simulations. Particularly, the GST-based approach provides a balanced way for cancellous bone anisotropic mechanical behavior prediction with low resolution clinical images, which can be also further applied to other porous materials.