The field of 3D bioprinting is dynamically advancing, witnessing constant evolution through the introduction of novel materials, techniques, and strategies. A persistent challenge within this domain pertains to the optimization of materials and processes, aiming to achieve both a high cellular viability in the printed structure incorporating cells and a faithful reproduction of the desired structure. The primary obstacle often encountered is the formulation of an appropriate bioink composition, exacerbated by the inherent variability in fluidic, rheological, and biocompatible characteristics. In this study, we investigated collagen-based bioinks crosslinked with alginates and reinforced with Laponite-RD nanoparticles. This formulation was explored as a potential yield-stress bioink, specifically addressing the dual challenge of ensuring high cellular viability and maintaining structural fidelity during bioprinting. Employing a Design of Experiment (DoE) methodology, we systematically gathered experimental data by varying material compositions. Subsequently, we assessed printability, extrudability, and viscoelastic properties. The acquired data were utilized to train a hyperparameter-tuned machine learning (ML) model, enabling the prediction of inter-relationships among the responses. Furthermore, the ML model was instrumental in identifying the optimal compositional formulation. This formulation facilitated the fabrication of a high-fidelity structure with suitable extrudability, mitigating shear-induced cell death and ensuring high cell viability in the post-printed structure. In summary, our findings underscore the efficacy of a data-driven experimental design coupled with ML methodologies in shaping the future landscape of high-resolution and high-viability 3D bioprinting. This synergistic approach holds promise for overcoming existing challenges and advancing the frontier of precision in Bio-Fabrication.
Qavi et al. (Tue,) studied this question.