Computational Mechanobiology in Cartilage and Bone Tissue Engineering: From Cell Phenotype to Tissue Structure - Computational Modeling in Tissue Engineering - page 354

A multiscale approach to scaffold design spanning a number of length scales
and hierarchies has been presented in Chan et al. [ 18 ] for hydroxyapatite-collagen
composite materials. The elastic properties of the HA nanoparticles were deter-
mined from first (i.e. quantum mechanical) principles. The constitutive behaviour
of the HA-collagen composite was then determined for various volume ratios
using microstructural 2D unit cell FE modelling of a HA particle embedded in a
collagen matrix. The collected data was used to investigate the behaviour of a 3D
RVE of the scaffold architecture in order to optimise pore size, pore density and
HA volume fraction for cortical and cancellous bone. While the properties of
cancellous bone could be matched, those of cortical bone could not be reproduced.
Layer-by-layer scaffold fabrication based on printing techniques was envisioned to
manufacture the computationally optimised scaffold.
A middle-out multiscale approach to assess fracture risk in a proximal femur
and an outline for appropriate verification and validation methods was introduced
by Cristofolini et al. [ 24 ] spanning several scales from the body to the cell level.
The central part of the approach is the organ model of the bone where the fracture
occurs. This model and its boundary conditions are informed by the body level
model establishing the musculoskeletal loads during daily activity. Downstream,
the organ level model also relies on the tissue level model where constitutive
relations are used to derive stresses and strains as well as the risk for material
failure (fracture). The tissue level model is coupled to the cell level model, that
provides the material properties for the tissue level model and relies on the input of
biophysical stimuli from the tissue level model to simulate bone cell activity and
remodelling. This type of hierarchical approach potentially allows the investiga-
tion of a multitude of factors as diverse as pharmacological treatment and subject
specific gait patterns and activity levels on the fracture risk of a bone. Currently
however, many questions regarding the individual component models remain
unresolved and under investigation. For any clinical use the various levels of the
model also need to be coupled in a fully automated and robust fashion to improve
usability.
5.2 Tissue Structure in Multiscale Simulations of Tissue
Regeneration
Bone and cartilage are anisotropic tissues and as such scaffolds can provide
environmental cues that guide subsequent anisotropic tissue formation [ 30 , 47 ].
Microscopic tissue structure has been incorporated into models investigating the
biomechanics of tissue behaviour. Sander et al. [ 98 ] report a recent example in the
context of tissue equivalent mechanics. Based on orientation and alignment
measured using polarised light imaging unique 3D fibre networks were created in
each point of the cell-compacted collagen gels. Based on constitutive parameters
defined at the fibre level and fit to macroscopic off-axis hold tests, the model was