Partenaires

Logo tutelle
CNRS



Rechercher

Sur ce site

Sur le Web du CNRS


Accueil du site > Équipes > Équipe 1 : Méthodes de caractérisation et d’imagerie ultrasonore multi-echelle : modélisation et transfert > Thèmes de recherche > Modélisation mécanique de l’os cortical.

Modélisation mécanique de l’os cortical.

Contacts : Quentin Grimal, Maryline Talmant.

Mechanical models of cortical bone elastic properties are a keystone to different ultrasonic applications developed in the laboratory. Their purpose is to relate some nanoscopic and microscopic features of bone tissue to its overall elastic properties, based on the principles of continuum mechanics. Models of bone tissue are multi-scale to account for the hierarchical organization of bone. They can be used (i) to relate the characteristics of the nanoscopic constituents of bone (mineral particles and collagen molecules) to the lamellar-level elastic properties, or (ii) to relate the porous network and the lamellar-level tissue elasticity to millimetre-scale properties. Realistic models of bone elastic properties must account for the organization of the bone matrix (orientation and content of mineral, quality of collagen) and the shape of the porous network. Elastic properties are anisotropic at different scales as a result of the composition and organisation of bone. Models are developed and validated based on experimental data obtained from multimodal experiments (ultrasound, mechanical testing, microtomography, etc.). Bone tissue models are helpful to set the bone mechanical properties in the computation codes of the ultrasonic propagation at the macroscopic scale, i.e., for the simulation of the QUS acoustic problems in vivo. Mechanical models can be run systematically to investigate the dependence of the various elastic constants on small-scales features (mineral content and orientation, porosity, etc.). When it comes to solving inverse problems based on in vivo ultrasound data, mechanical models can be used as a priori information to diminish the number of unknowns or define appropriate regularisation of the inverse problem algorithm.