Direct ink writing of β-TCP/Pluronic F-127 scaffolds: Effect of printing parameters on line width

Osteoporosis is a medical condition in which the bones of the patients got weaken resulting in fracture of bones. Tissue engineering has emerged to address this condition by employing scaffolds at the damaged part for repairing and regenerating damaged bone tissues. The ideal scaffolds should be of biomaterials with porous structure for facilitating nutrient diffusion and mechanical support. Fabrication of this porous scaffolds required high precision techniques and suitable biocompatible regenerative materials. Direct ink writing (DIW) is a versatile extrusion based Additive Manufacturing (AM) technique enables the fabrication of scaffolds with tailored porous architecture mimicking the natural bones. This work studies the effect of printing parameters on the line width. Box–Behnken design was employed for the experimental model, ANOVA for statistical analysis, and a machine learning (ML) based approach to identify the optimal printing parameters. Multiple regression models were evaluated, including Linear Regression, Random Forest, Gradient Boosting, and XGBoost. The Linear Regression demonstrating the best performance (RMSE: 0.1147, R²: 0.7204) a result attributable to the small, near-linear dataset rather than a demonstrated advantage of ensemble methods over the quadratic RSM model (R²: 0.9836). Further, scaffolds were fabricated using the optimized conditions to assess the shrinkage, compressive strength, density, porosity, phase composition (XRD), and pore-size distribution of the bioceramic scaffold made of β-Tricalcium Phosphate (β-TCP) prior to sintering. The properties of the green (pre-sintering) printed scaffolds indicate promising potential for bone tissue engineering applications, pending post-sintering and biological evaluation.