Replicating trabecular bone through 3D printing

Abstract

Background: 3D printing, particularly fused deposition modelling, has become a valuable tool in medical imaging and radiotherapy for creating anthropomorphic phantoms. These models are cost-effective and patient-specific, offering high anatomical and radiological accuracy. While cortical bone replication has been explored extensively, trabecular bone simulation remains less developed and presents additional challenges due to its complex internal structure. Objectives: This research aims to develop a method for replicating both trabecular and cortical bone to produce anatomically and radiologically accurate patient-specific bone phantoms. The approach involves varying the infill density and optimising material selection and printing parameters to achieve Hounsfield unit values equivalent to those of real bone. Methodology: Five composite filaments were used to 3D print cylindrical test specimens with varying infill densities and patterns. Computed tomography scans measured their HU values, guiding the optimisation of print settings to closely match the HU values of human bone. Anonymised femur CT data was then used to fabricate an anatomically accurate femur phantom using the optimised materials and configurations. Results: Iron PLA closely matched cortical bone, achieving HU values up to 1754.7 at 70% rectilinear infill and as low as 159.8 at 25% infill. PLA Stone was most suitable for trabecular bone, with values ranging from 83.2 at 60% gyroid infill to 865.6 at 99% cubic infill. The final phantom used PLA Stone at 70% gyroid infill for the head and distal epiphysis, and Iron PLA at 60% rectilinear infill for the cortical shaft, accurately replicating both the anatomy and radiodensity. Conclusions: The study demonstrated a reliable, low-cost method for creating patient specific bone phantoms using fused deposition modelling. These phantoms have strong potential to enhance clinical quality assurance, treatment planning, and educational applications, offering an accessible alternative to traditional phantoms. Recommendations: For medical physics practice, it is recommended to establish QA protocols using these phantoms and consider setting up in-house 3D printing to support clinical and educational applications. For future research, further work could explore advanced printing techniques such as dual-extrusion fused deposition modelling and selective laser sintering, expand phantom development to other anatomical structures.