Progress in multiscale biomechanical assessment of cancellous bone: From bone density to multiscale mechanical...

Cancellous bone is a porous tissue formed by a network of plate- and rod-like trabeculae. Its mechanical properties are determined by factors across multiple scales, including bone mass, the spatial architecture of the trabeculae, and tissue composition. The clinically standard measure, Bone Mineral Density (BMD), only reflects mineral content per unit volume. It cannot fully characterize bone quality, which encompasses trabecular microstructure, degree of mineralization, and collagen characteristics. This limitation results in a fracture prediction accuracy of only about 60%. Consequently, assessment methods based on multiscale mechanical phenotypes have emerged. These approaches integrate evaluations from the macro-scale (overall mechanical properties), meso-scale (trabecular network), and micro-/nano-scale (tissue composition and local mechanics). This comprehensive strategy provides a more accurate representation of the true mechanical state of cancellous bone. This review systematically summarizes recent progress in the multiscale biomechanical assessment of cancellous bone. It covers evaluation techniques at three levels: the macro-scale (e.g., Quantitative Computed Tomography-QCT, Biomechanical Computed Tomography-BCT, Finite Element Analysis - FEA); the meso-scale (e.g., High-Resolution Peripheral Quantitative Computed Tomography-HR-pQCT, High-Resolution Magnetic Resonance Imaging-HR-MRI); and the micro-/nano-scale (e.g., Raman spectroscopic imaging-RSI, Ultrashort Echo Time Magnetic Resonance-UTE-MR, nanoindentation). For each, we analyze their respective advantages and limitations. Furthermore, we explore the potential applications of these multiscale approaches in fracture risk assessment, the design of patient-specific implants, and research into the mechanisms of bone metabolic diseases. We distinguish two pathways for clinical translation: the personalized configuration of standard implants, involving preoperative planning and intraoperative adjustment of off-the-shelf devices guided by the patient's mechanical phenotype, and the design of fully customized, patient-specific implants for complex bone defects. Within this framework, we discusses the potential in three areas: fracture risk assessment, personalized surgeries planning, and mechanistic studies of bone metabolic diseases-and conclude with future directions for artificial intelligence-driven multiscale data integration.