Computational optimization of 3D printed bone scaffolds using orthogonal array-driven FEA and neural network modeling

Area of Science:

• Biomaterials Science
• Orthopedic Engineering
• Tissue Engineering

Background:

• Bone tissue engineering seeks to regenerate damaged bone using biomimetic scaffolds and advanced materials.
• Optimizing scaffold design is crucial for mechanical integrity and biological performance in bone regeneration.

Purpose of the Study:

• To optimize 3D printed polylactic acid (PLA) lattice scaffolds for bone tissue engineering applications.
• To investigate the influence of geometric configuration and processing parameters on scaffold mechanical performance.

Main Methods:

• Development of Lidinoid, Diamond, and Gyroid lattice scaffolds with varying wall thicknesses (1.0-2.0 mm).
• Application of Taguchi L27 Orthogonal Array for experimental design and mechanical testing under compressive loads (3-9 kN).
• Utilized Back-propagation Artificial Neural Network (BPANN) for predictive modeling and Finite Element Analysis (FEA) for validation.

Main Results:

• The Gyroid lattice with 2.0 mm thickness demonstrated superior mechanical integrity, exhibiting minimal displacement (0.36 mm) and strain (1.2 x 10⁻²) at 3 kN.
• Lidinoid structures showed the highest deformability under tested conditions.
• BPANN model achieved high prediction accuracy (R² > 0.995) for scaffold displacement and strain.

Conclusions:

• The Gyroid lattice geometry represents a promising scaffold design for bone tissue engineering due to its enhanced mechanical properties.
• An integrated approach combining experimental design, machine learning, and FEA provides a robust framework for optimizing scaffold architecture.
• Optimized scaffolds have significant implications for improving mechanical strength and biological outcomes in bone healing applications.