Towards Personalized 3D Bioprinting: Leveraging Cluster-based Machine Learning for Mass Customization

3D bioprinting emerged as a potential technology for biomedical and tissue engineering. Mass customization in terms of material selection, cell types, fabrication techniques, along with shortened fabrication time made 3D bioprinting very attractive in medical-related applications. However, the bioprinting process remains challenging in two folds: finding optimal process parameters for desired mechanical properties of scaffolds and obtaining cell type-based microenvironment into fabricated scaffolds. Getting both the personalized mechanical properties and microenvironment of 3D bioprinted scaffolds simultaneously sometimes even inversely related. In this work, we adopted the clustering-based machine learning technique to understand both the mechanical and microenvironment of scaffolds. K-means clustering would apply to obtain an optimal number of clusters for the bioprinting experimental dataset. Each cluster obtained from cluster-based machine learning technique will be analyzed and rated based on the patterns toward bioprinting of scaffold ensuring user defined mechanical properties and personalized microenvironment. This new approach would help 3D bioprinting practitioners to address the dual challenge of achieving desired properties and microenvironment in printed scaffolds.