3D printing of lung nodules

Some examples of the 3D printed lung nodules (resin, PA12, TPU), both geoemetric and segmented from patients, are shown in Figure 3

Fig 3: 3D printed nodules based on geometric shapes (orange, resin PRUSA), gray (Polyjet, MJF), white SLS PA11. Some were printed inside grids to save costs and guarantee that they were printed simultaneously.

Fig 4: Models (left) of lung nodules together with their 3D printed versions, central slice in CT and segmented model from CT images (right).

Evaluation of attenuation and morphometry of 3D-printed lung nodules: CT imaging

The CT values (Average±SD) for the selected materials (solid and subsolid nodules) were: [MJF:TPU(60±15HU); PA11(-53±30HU);PA12(-100±45HU);PP(-180±25HU)], [SLS:PA11(-55 ±15HU);PA12(-80 ±20HU)]; [PRUSA:orange-resin(69±20HU);grey-resin(90±32HU)]. For GGOs, the subsolid component was in the range (-630,-550HU). All are in the range of commercially available lung nodules and patient nodules data. 

Figure 5 shows the CT values of the different nodules and materials evaluated and their equivalence to commercial nodules.

Fig 5: CT attenuation of 3D printed nodules in a box-whisker plot representing all the nodules geometries evaluated.

Morphometry analysis

3D-printed nodules showed satisfactory accuracy compared to the design models with volume differences<5% and average absolute distances<0.6mm.

Figure 6,

Fig 6: Morphometry analysis of 3d printed nodules based on patient data segmented from CT images and compared to original modules. The coloured map represents the signed distance between model and 3D printed nodule.

Combination of lung nodules and anthropomorphic commercial and 3D-printed lung phantoms: CT imaging

Two examples of 3D printed nodules' setups combined with the 3D printed custom lung vessel phantoms and the commercial phantoms are shown in Figure 7

Fig 7: Setup of 3D printed lung nodules combined with a custom 3D printed thorax phantom (left) and a commercial lung phantom (right).

The appearance of the lung nodules in CT images was realistic, as illustrated when combined with the 3D printed lung vessel custom phantom (Fig. 8 and 9). In these images, the powder pockets appear mimicking ground glass opacities. Other geometric and patient based nodules can be seen as well

Fig 8: CT images of 3D printed left lung insert with patient-based, geometric and opacities (powder traps) in GIF form, thin slices (0.625mm) and lung window.

Fig 9: CT images of 3D printed right lung insert with patient-based, geometric and opacities (powder traps) in GIF form, thin slices (0.625mm) and lung window.

Two zoomed in images of one of the hollowed spherical nodules, containing powder to mimic opacities are visualized in Fig. 10 and 11. In the CT images with thin slice thickness the nodule edge is highlighted

Fig 10: Spherical nodule mimicking a GGO (thick slices with a lung window)

Fig 11: Spherical nodule mimicking a GGO (thin slices with a lung window), observe the thin wall (solid) higher attenuation

A selection of lung nodules, including a GGO placed in the commercial lung phantom are illustrated in Fig. 12 and 13

Fig 12: CT slice of the Kyoto Kagaku thorax paediatric phantom showing several 3D printed lung nodules

Fig 13: CT slice of the Kyoto Kagaku thorax paediatric phantom showing several 3D printed lung nodules at a different location