Translating in silico micro-computed tomography bone scans into a 3D printable bone model

Abstract

Personalised medicine is an expanding area of medical research, in which patient-specific physiology informs clinical interventions. The personalisation of skeletal-based medical implants through 3D imaging and additive manufacturing (AM) has the potential to improve patient outcomes and reduce the likelihood of implant rejections. Currently, full-body computed tomography (CT) scanners obtain images at the limit of resolution of intricate internal bone architecture such as trabecular bone (Tb), which makes them difficult to map with precision in humans. Therefore, we have proposed the development of a 3D printing workflow which utilises scaled high-resolution micro-CT outputs that maintain bone topography in the final 3D print, specifically the retention of physiologically relevant trabeculae. In this study, we evaluate 3D model generation and rendering parameters that are necessary to maintain anatomical fidelity while ensuring printability. Meshing algorithm, computer-aided design (CAD) rendering tools, and the resulting surface tessellation of ex vivo bone samples were evaluated to identify computational limitations and their impact on downstream fabrication. A proof-of-concept workflow was developed, in which high-resolution micro-CT datasets preserving intact Tb morphology were successfully scaled and translated into anatomically informed 3D-printed models using fused deposition modelling (FDM) and stereolithography (SLA).