Fig 1.
Skull base regions of interest.
A: View of the left orbital lamina which is a smooth and oblong bone plate, part of the lateral surface of the ethmoid bone. B: Endocranial view of the body of the sphenoid bone.
Fig 2.
Main steps for geometric and mechanical evaluation of materials.
Five human cadaverous heads were used: first one was a control head requiring medical scanner (CT) acquisition only; 2 were dissected to perform bone samples and acquired in CT and microCT scanner (μCT), then 3D-printed; the last two were only dissected to perform bone samples. 3D-printed samples were in turn acquired in CT. Two types of experiments were then realized: geometric tests on 3D computer reconstruction of bone samples acquired in CT and μCT, compared with those of 3D-printed samples; mechanical tests on bone samples and 3-D printed samples.
Fig 3.
Illustration of imaging accuracy.
A. Photography of an orbital lamina, B. CT-mesh, C. μCT-mesh.
Table 1.
Table summarize material and printer properties as specified by manufacturers.
Fig 4.
Main mechanical and geometric tests results.
A: Absolute distance between material mesh compared to bone mesh (boxplot). Distance error between material and bone landmarks (brown points). B: Bias evaluation: accuracy of CT compared with μCT and smoothing steps. C: Forces required to break thin walls (orbital lamina). D: Forces required to break thin walls (anterior wall of the sphenoidal sinus). E: Comparison of energy spent in a 6mm depth drilling on cadaver bone and materials. F: Repartition of instantaneous forces applied on materials during a 48mm depth cumulated drilling.
Fig 5.
Macroscopic aspect of broken materials (orbital lamina).
Red point represents the pressure point. A: Sample positioned on a circular support for the breaking test. B: Cadaver bone, C: PC: the fracture mode is similar to the cadaver bone, D: Multicolor: the fracture line is distant to the pressure point.
Table 2.
Table summarize material properties regarding initial specifications.
Fig 6.
A. View of the installation for surgical training. B. View of the entire 3D-printed model (on the left) and internal details of ethmoidal and sphenoidal sinuses (on the right). C. Endoscopic views of the model and training procedures: resection of ethmoidal cells with a rongeur and breaking walls with a suction tip.