Page 21 - IJB-5-1

Additive manufacturing of bone scaffolds
exhibits poor controllability on the structural performance [48] optimized a scaffold based on the SIMP optimization
and mechanical properties of designed scaffolds. algorithm, in which two competing properties, including
Moreover, there is an ineffaceable staircase phenomenon the modulus and permeability, were tailored using a
on the external contour of models caused by Boolean single objective function . SIMP method is also applied
[49]
operation, leading to geometric distortion and mechanical to optimize the elastic tensor of scaffolds . In vivo tests
[50]
instability . Besides, the CAD-based method can only showed that the optimized scaffolds with similar elastic
[42]
design the scaffolds with periodic and regular structure. properties to that of human bone exhibited an accelerated
bone remodeling rate. Besides, a series of unit cells with
2.2. Topology Optimization maximal shear and bulk modulus, predefined stiffness
Ideal bone scaffolds should not only have highly porous ratios and functionally graded structure, were obtained
[51]
structure to facilitate cell in-growth and nutrient transport through the ESO-based topology optimization . Some
but should also possess enough mechanical properties scaffolds optimized by bidirectional evolutionary ESO
to provide stable structure support [30,43-45] . Paradoxically, method, which obtain the maximum bulk or shear
increasing the porosity of the scaffolds enhances the modulus under various prescribed volume fractions, are
[52]
material transport capacity but inevitably impairs the depicted in Figure 3 .
mechanical properties. Thus, the scaffold designer should Another typical topology optimization method is
balance these two conflicting properties to obtain an optimal level-set al.orithm, which centers on tracing the
comprehensive performance. Topology optimization is phase boundaries, thus effectively describing smooth
[53]
a method to optimize the distribution of materials in a boundaries to control the topology changes . A level-
given region based on the given load condition, constraint set based method for scaffolds design was proposed to
condition, and performance index . Designing scaffolds obtain material with maximal permeability . Level-
[54]
[46]
with topology optimization is expected to achieve an set based topology optimization made it possible that
optimized comprehensive performance with certain the no-slip boundary condition of fluids in Stokes flow
constraints, for example, the maximum mechanical could be naturally satisfied . Topology optimization
[55]
properties with certain porosity or maximum permeability. was also reported to tailor the thermal expansion of
In topology optimization of scaffolds, the optimization porous multimaterials . However, as mentioned before,
[56]
problem is generally solved indirectly through optimizing the topology optimization of the scaffolds begins with
a unit cell with specific optimization algorithms. After the optimization of the unit cells and proceeds with the
obtaining the optimized unit cell architecture, the whole subsequent periodic arrangement. Therefore, it can only
scaffold is formed by repeating it periodically. The classic achieve regular porous architecture, which is considerably
optimization algorithms applied in topology optimization different from the irregular structure of natural bone.
of scaffolds include Solid Isotropic Material with
Penalization (SIMP) method and evolutionary structural 2.3. Reverse Modeling
optimization (ESO) method, which describe the structure Reverse modeling design, also known as image-based
point-by-point in topology optimization . Guest et al. design, reconstructs bone tissue microstructure directly
[47]
based on object’s computed tomography (CT) or magnetic
resonance imaging (MRI) . In this method, the CT/MRI
[57]
A
slice images undergo a series of analysis, with an aim to
extract the key features for reconstruction. Binary value
method is commonly used to analyze the slice information,
in which element “1” represents the solid, whereas “0”
represents the void . Then, the pre-defined unit cell is
[58]
mapped according to the extracted slice information to
construct the 2D model. This 2D model will directly be
B transformed into STL files and transmitted to an AM
equipment to construct 2D layer. After this, a 3D part
is obtained using the layer-by-layer method. Obviously,
reverse modeling design combines advanced medical
imaging system, powerful image analysis software as
well as rapid AM technique, which guarantees a more
mimic architecture for bone tissue engineering.
Figure 3. Topology optimized unit cells and scaffolds with (A) Reverse modeling has appealed to a large number of
maximum bulk modulus and (B) maximum shear modulus under researchers for constructing customized scaffolds. Sun et al.
various predefined volume fractions . [59] systematically investigated the modeling principles and
[52]
4 International Journal of Bioprinting (2019)–Volume 5, Issue 1