Optimizing finite element predictions of local subchondral bone structural stiffness using neural network-derived density-modulus relationships for proximal tibial subchondral cortical and trabecular bone
Modeled proximal tibial structural stiffness using finite element modeling Validated stiffness predictions using in situ macro-indentation testing Applied a neural network to derive optimized modeling parameters Finite element modeling explained 81% variance in structural stiffness Require de-blurring methods, anisotropy inclusion and a smaller voxel to improve predictions