Characterization and optimization of a biomaterial ink aided by machine learning-assisted parameter suggestion

Abstract

Bio-inks and biomaterial inks are crucial to the success of 3D bioprinting, as they form the foundation of almost every 3D bio-printed structure. Despite the use of various biomaterial inks with potential biomedical applications in 3D printing, developing printable biomaterial inks for extrusion-based 3D bioprinting remains a major challenge in additive manufacturing. To be effective, the inks must possess suitable mechanical properties, high biocompatibility, and the ability to print precisely. In this study, machine learning (ML) was employed to develop a chitosan-gelatin-agarose biomaterial ink. The ink's printability, rheological properties, hydrophilicity, degradability, and biological response were evaluated after an optimization process. The optimized ink exhibited adequate viscosity for reliable printing, and 3D structures were created to assess printability and shape integrity. Bone marrow mesenchymal stem/stromal cells (BMSCs) were cultured on the ink's surface, and cell adhesion, growth, and morphology were assessed. Results showed favorable cell morphology, and cell viability within the optimized ink. The ink consisting of 27 % agarose, 53 % chitosan, and 20 % gelatin (ACG), may be a suitable biomaterial for fabricating 3D complex tissue constructs.