On the optimization of low-cost FDM 3D printers for accurate replication of patient-specific abdominal aortic aneurysm geometry

Abstract

Background: There is a potential for direct model manufacturing of abdominal aortic aneurysm (AAA) using 3D printing technique for generating flexible semi-transparent prototypes. A patient-specific AAA model was manufactured using fused deposition modelling (FDM) 3D printing technology. A flexible, semi-transparent thermoplastic polyurethane (TPU), called Cheetah Water (produced by Ninjatek, USA), was used as the flexible, transparent material for model manufacture with a hydrophilic support structure 3D printed with polyvinyl alcohol (PVA). Printing parameters were investigated to evaluate their effect on 3D-printing precision and transparency of the final model. ISO standard tear resistance tests were carried out on Ninjatek Cheetah specimens for a comparison of tear strength with silicone rubbers.

Results: It was found that an increase in printing speed decreased printing accuracy, whilst using an infill percentage of 100% and printing nozzle temperature of 255 °C produced the most transparent results. The model had fair transparency, allowing external inspection of model inserts such as stent grafts, and good flexibility with an overall discrepancy between CAD and physical model average wall thicknesses of 0.05 mm (2.5% thicker than the CAD model). The tear resistance test found Ninjatek Cheetah TPU to have an average tear resistance of 83 kN/m, higher than any of the silicone rubbers used in previous AAA model manufacture. The model had lower cost (4.50 GBP per model), shorter manufacturing time (25 h 3 min) and an acceptable level of accuracy (2.61% error) compared to other methods.

Conclusions: It was concluded that the model would be of use in endovascular aneurysm repair planning and education, particularly for practicing placement of hooked or barbed stents, due to the model's balance of flexibility, transparency, robustness and cost-effectiveness.

Keywords: 3D printing; Abdominal aortic aneurysms; Accurate; Flexible; Rapid prototype; Thermoplastic polyurethane; Transparent.

Fig. 1

Picture of the 3D reconstruction from the patient’s AAA geometry. The picture was obtained from CT scan data, which was transferred to the software Cura (version 2.5, Ultimaker, Geldermalsen, Netherlands) for layer slicing and setting the 3D printing parameters

Fig. 2

3D printed Ninjatek Cheetah Water TPU angle test piece that complies with ISO 34–1:2015 standards. This specimen was used for tear resistance test measurements with a 1 mm nick at point A

Fig. 3

3D printed rings with 2 mm wall thickness for the application of AAA modelling. a and b The 3D printed rings are flexible. c Rings printed with 30, 40, 50 and 60 mm/s printing speeds (see Additional file 1: Table S7 in the Appendix for the rest of the printing parameters) had fair transparency and a noticeable join line that was constant throughout all prints

Fig. 4

The infill density changes the sample transparency. The 100% infill density displays better transparency than the 0% infill density. The 20 and 50% infill densities have partial layer fill, clouding the transparency. With 100% infill the sample is has no air gaps; thus, no clouding of optical properties appears

Fig. 5

3D printing temperature effects on the transparency. With increasing printing nozzle temperature the transparency increases

Fig. 6

a AAA model printed with PVA support structure. b model after PVA support was dissolved in warm water bath. A black object was placed inside to display level of transparency

Fig. 7

Printed AAA model with cut sections for use in accuracy testing

Fig. 8

Ninjatek Cheetah tear strength tests for nicked angle test specimens according ISO standard 34–1:2015