EVALUATION OF THE ACCURACY OF DENTAL MODELS FABRICATED BY VARIOUS 3D PRINTERS

Abstract

the mean deviations gradually decreased as the position shifted rightward. In contrast, SLA and Polyjet showed the most negative deviations in the left second molar position in the X-axis and values gradually increased as the position shifted rightward. In the Y-axis, FDM exhibited forward deviations on the right side and DLP showed backward deviations in the anterior site and forward deviations in the posterior site. In the Z-axis, all printers showed little deviation in canine and premolar regions, with maximum deviation in second molar sites. The total linear deviation was lowest in the left canine region and deviations increased with distance from this site in all groups. Conclusion. FDM showed more systemic deviations than DLP, Polyjet and SLA in the superimposition study, but all printed models were within a clinically acceptable range. Bucco-lingually, FDM and DLP showed expansion while Polyjet and SLA showed contraction. Anterior-posteriorly, FDM generally showed forward deviations and DLP showed forward deviations in the molar region and backward deviations in the anterior region. In the vertical dimension, there were no significant deviations in the anterior region but significant deviations were found in the molar region. For all printers, total linear deviations were lowest in the anterior region and became larger as the position moved backwards.

ABSTRACT EVALUATION OF THE ACCURACY OF DENTAL MODELS FABRICATED BY VARIOUS 3D PRINTERS Jin Jeon, D.D.S., M.S. Department of Prosthodontics, Graduate School, Seoul National University (Directed by Professor Seong-Joo Heo, D.D.S., M.S.D., Ph.D.) Objectives. Optical impressions obtained with an intraoral scanner and prosthesis fabrication using computer-aided design (CAD)/computer-aided manufacturing (CAD) were introduced into the field of dentistry in the early 1980s. With the recent development and increasing popularity of intraoral scanners, direct digital acquisition of the dental arch in the mouth has become more readily available, omitting the step of conventional impression acquisition. In some simple cases, physical models are not required with digital impressions

however, physical dental models are mandatory in many extensive cases. Rapid prototyping or 3D printing can be used to fabricate models from digital impression data. The purpose of this study was to evaluate the dimensional accuracy and reproducibility of compartments in full arch models made with various 3D printing methods including digital light processing (DLP), fused deposition modeling (FDM), stereolithography (SLP) and photopolymer jetting (Polyjet). Material and Methods. The mandibular model was modified to generate reference CAD data in reverse engineering software. To simulate an implant scan body, 6 cylinders were placed in the left and right canines (#33, #43), second premolars (#35, #45) and second molars (#37, #47). The cylinders in the left (#37) and right second molars (#47) were designed to have mesial and distal angulation of 30° to simulate off-axis implant angulations. Three additional reference spheres were placed around the left second molar (#37) to set coordinates for deviations. Ten models were printed with 4 different printers and 5 materials using the CAD data file. The printed models were scanned with a model scanner and merged with the original CAD data. Difference color maps were used to evaluate deviations and surface textures were examined with photographs and scanning electron microscopy. The systemic mean deviations of models from each 3D printer were analyzed in a superimposition study. Deviations at different cylinder locations in the X-, Y-, and Z-axes, and total linear deviations were analyzed in a coordinate study. Results. In the superimposition study, FDM yielded the highest mean deviations with statistical significance (p < 0.05). There were no significant differences among the other printer groups. In the coordinate study, FDM and DLP showed the most positive deviations in the left second molar position in the X-axis