Abstract
Among various 3D bioprinting methods, extrusion-based bioprinting stands out for its ability to maintain high cell viability and create intricate scaffold structures. However, working with synthetic polymers or natural shear-thinning hydrogels requires precise control of rheological properties, such as viscosity, to ensure scaffold stability while supporting living cells. Traditionally, researchers address these challenges through extensive experimentation, separately optimizing material properties and bioprinting performance. This process, though effective, is often slow and resource-heavy. To streamline this workflow, computational approaches like machine learning are proving invaluable. In this study, a decision tree model was developed to predict the viscosity of bioinks across various compositions with high accuracy, significantly reducing the trial-and-error phase of experimentation. Once viscosity is optimized, k-means clustering is applied to analyze and group scaffolds based on their mechanical and biological properties. This clustering technique identifies the optimal characteristics for scaffolds, balancing structural fidelity and cell viability. The integration of these computational tools allows researchers to optimize bioink formulations and printing parameters more efficiently. By reducing experimental workload and improving precision, this approach not only accelerates the bioprinting process but also ensures that the resulting scaffolds meet the required mechanical integrity and provide a conducive environment for cell growth. This study represents a significant step forward in tissue engineering, offering a robust, data-driven pathway to enhance both the efficiency and quality of 3D bioprinted constructs.