Date Published: August 13, 2018
Publisher: Springer International Publishing
Author(s): Takashi Kamio, Kamichika Hayashi, Takeshi Onda, Takashi Takaki, Takahiko Shibahara, Takashi Yakushiji, Takeo Shibui, Hiroshi Kato.
In the oral and maxillofacial surgery and dentistry fields, the use of three-dimensional (3D) patient-specific organ models is increasing, which has increased the cost of obtaining them. We developed an environment in our facility in which we can design, fabricate, and use 3D models called the “One-stop 3D printing lab”. The lab made it possible to quickly and inexpensively produce the 3D models that are indispensable for oral and maxillofacial surgery. We report our 3D model fabrication environment after determining the dimensional accuracy of the models with different laminating pitches (; layer thickness) after fabricating over 300 3D models. Considerations were made for further reducing modeling cost and model print time. MDCT imaging was performed using a dry human mandible, and 3D CAD data were generated from the DICOM image data. 3D models were fabricated with a fused deposition modeling (FDM) 3D printer MF-2000 (MUTOH) with a laminating pitch of 0.2 mm, 0.3 mm, 0.4 mm, or 0.5 mm. Each 3D model was then subjected to reverse scanning to evaluate the modeling conditions and deformation during modeling. For the 3D image processing system, Volume Extractor 3.0 (i-Plants Systems) and POLYGONALmeister V2 (UEL) were used. For the comparative evaluation of CAD data, spGauge 2014.1 (Armonicos) was used.
As the laminating pitch increased, the weight of the 3D model, model print time, and material cost decreased, and no significant reduction in geometric accuracy was observed.
The amount of modeling material used and preparation cost were reduced by increasing the laminating pitch. The “One-stop 3D printing lab” made it possible to produce 3D models daily. The use of 3D models in the oral and maxillofacial surgery and dentistry fields will likely increase, and we expect that low-cost FDM 3D printers that can produce low-cost 3D models will play a significant role.
Three-dimensional (3D) patient-specific organ models made with 3D printing technology are utilized in various fields [1–4]. In the oral and maxillofacial surgery and dentistry fields, 3D models of hard tissues such as teeth and bones are being utilized for medical education training, explanation to the patient, operation planning, and simulated surgery using real surgical instruments [5–8]. The increased use of 3D models has directly led to an increase in the cost of obtaining them. Reducing the cost of obtaining 3D models is now one of the major concerns. By generalizing the hardware and software surrounding 3D printing technology [9, 10], a desktop fused deposition modeling (FDM) 3D printer which is extremely inexpensive compared with industrial 3D printers, we created an environment for enabling design, fabrication, and the use of patient-specific 3D models in our facility entitled the “One-stop 3D printing lab”. 3D models were produced quickly and the cost burden was greatly reduced. The laminating pitch (layer thickness) and fill density (infill density) control the amount of modeling material used. While it is expected that an increase in laminating pitch will lead to a reduction in the modeling cost, there is concern that the precision will be lowered.
The shape error between the CAD model and the printed 3D object was measured to understand the printing characteristics. MDCT scanning was performed on the dry human mandibular bone, and then 3D CAD data in the STL format file (composed of about 100,000-point clouds) were created from the DICOM image data. Medical image processing software was used to create CAD data, and a desktop FDM 3D printer was used to fabricate the 3D model (Fig. 1). 3D models with laminating pitches of 0.2 mm, 0.3 mm, 0.4 mm, or 0.5 mm were created from the original 3D mandibular bone CAD data (Figs. 2a–d and 3a). MDCT reverse scanning of each fabricated 3D model was performed under the same conditions. The shape error of each 3D model against the 3D CAD model was calculated and the dimensional accuracy was evaluated.Fig. 1The FDM 3D printer, Value3D MagiX MF-2000Fig. 23D models with laminating pitches of 0.2 mm (a), 0.3 mm (b), 0.4 mm (c), and 0.5 mm (d)Fig. 3Visualization of shape error (signed differences) for each 3D CAD model. Warm color shows expansion rather than reference 3D CAD data, cold color shows shrinkage. a Reference 3D CAD data. b–e Slight changes in dimension were considered to be due to its own weight (arrowheads)
The results are shown in Table 1 and graphically in Figs. 3b, to e, 4a and b. The mean absolute shape errors of the laminating pitches of 0.2 mm, 0.3 mm, 0.4 mm, and 0.5 mm were 0.36 mm, 0.36 mm, 0.35 mm, and 0.35 mm, respectively. In the visualization of the shape error of each 3D CAD model, it is recognized that slight changes in the dimension occurred because of the own weight of the model. In particular, the tendency was found in the region of the lower edge and mandibular angles. As the laminating pitch increased, no significant reduction in geometric accuracy was observed.Table 1Outline of each fabricated 3D model and shape error evaluation with reference 3D CAD dataLaminating pitch0.2 mm0.3 mm0.4 mm0.5 mmModel print time4 h37 m3 h13 m2 h33 m2 h17 m3D model weight51 g50 g49 g48 gComparison with 3D CAD data Mean absolute shape error (mm)0.360.360.350.35 Minimum shape error (mm)−3.83−3.83− 3.78−3.93 Maximum shape error (mm)3.472.993.944.07 Standard deviation0.530.530.560.58Fig. 4a Signed shape error of each 3D CAD model. The solid black line represents median value. Top of the box (upper hinge) represents 75th percentile, and bottom of the box (lower hinge) represents 25th percentile. Whiskers represent maximum and minimum values. b Absolute unsigned shape error of each 3D CAD model. The solid black line represents median value. Top of the box (upper hinge) represents 75thpercentile, and bottom of the box (lower hinge) represents 25th percentile. Whiskers represent maximum and minimum values
Despite the expense, many facilities outsource their 3D modeling to external companies because of the work and time required for their creation. If inexpensively fabricating medical 3D models were to become possible, more needs could likely be met internally. The costs of the desktop 3D printer and the modeling materials are lower than those of professional 3D printers for industrial use. To promote the spread of 3D printers in the oral and maxillofacial surgery and dentistry fields, it is essential to accumulate knowledge about the modeling characteristics of 3D printers.
The results obtained using the FDM 3D printer suggested that adjusting the laminating pitch may lead to further reduction of model print time and cost. It was possible to quickly print a 3D model while greatly reducing the cost burden using the low-cost desktop 3D printer in the “One-stop 3D printing lab.”