3D printing has demonstrated extensive capability in fabricating patient-specific organs and tissues with significant design flexibility. In bone tissue engineering, 3D printing can be used to create customized scaffolds for treating large bone defects caused by injury or illness. More specifically, polymer-nanoceramic composites (PNCs) have shown effectiveness to mimic the properties of native bone tissue. However, the lack of geometric fidelity of the 3D-printed PNC scaffolds is a major challenge in their broader adoption in real-world bone tissue engineering applications. Although post-process characterizations can provide detailed information of build geometry, in-situ geometric fidelity characterization will facilitate real-time process adjustment and control. In this study, poly(lactic-co-glycolic acid) (PLGA) and nano-hydroxyapatite (nHA) were mixed and used as the PNC material due to the improved mechanical and biological properties they can jointly provide. In-situ layerwise images were captured and analyzed to characterize the geometric fidelity of the printed structure. More specifically, the pore sizes and strut widths over the entire layer were characterized as a tensor structure. Subsequently, a tensor-on-tensor regression model was used to quantify the relationship between the in-situ process features and ex-situ characterization of the print fidelity. The established machine learning model will facilitate real-time geometric fidelity prediction of PNC 3D printing. A case study based on the fabrication of a rectangular shaped scaffold was used to validate the proposed methodology. The sources of variability in the in-situ geometric fidelity prediction were discussed and some future research directions were summarized.

In-situ geometric fidelity characterization of 3D printed composite scaffolds for bone tissue engineering

ACS Spring 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.