Comparison of the accuracy (trueness and precision) of virtual dentofacial patients digitized by three different methods based on 3D facial and dental images

Abstract

Statement of problem: The accuracy of virtual dentofacial patients has been explored, but the accuracy of virtual patients established by using a straightforward and reliable method and the accuracy of different virtual patients are unclear.

Purpose: The purpose of this clinical study was to compare the accuracy of virtual dentofacial patients digitized by using registered-block impression, exposed anterior teeth, and cone beam computed tomography (CBCT) reconstruction methods based on 3-dimensional (3D) facial and dental images.

Material and methods: From the 15 selected participants who needed CBCT scanning, 3 kinds of virtual dentofacial patients were established by using 3 registration methods based on digital dental casts: 3D facial images, CBCT data, and registered-block impression. Compared with actual measurement, 25 linear distances of all virtual dentofacial patients were selected and measured by using a software program, and 3 separate measurements were calculated by the same person. The 1-way analysis of variance (ANOVA) was used to compare the deviations among 3 kinds of virtual dentofacial patients (trueness) and the deviations within groups (precision). The 1-sample t test was used to compare the difference between the deviation and the ideal error of 0.00 (α=.05).

Results: Compared with the actual measurement, the trueness of the average deviations for registered-block impression (1.02 ±1.24 mm) was better than that of exposed anterior teeth (2.35 ±1.71 mm) and CBCT reconstruction (2.86 ±1.61 mm). The precision of the average deviations for registered-block impression (1.29 ±1.43 mm) was better than that of exposed anterior teeth (2.00 ±1.72 mm) and CBCT reconstruction (2.12 ±1.94 mm). Significant differences in trueness and precision were found among the 3 groups of virtual dentofacial patients (P<.01). Significant differences among the deviations of all linear distances and the ideal error of 0.00 were observed for all groups of virtual dentofacial patients (P<.05).

Conclusions: The accuracy of registered-block impression was better than that of the exposed anterior teeth and CBCT reconstruction. The accuracy of exposed anterior teeth was lower than that of the other methods but could satisfy the requirements of clinical diagnostics and scientific methods. The accuracy of CBCT reconstruction was poor and could only be used for special situations that permitted low accuracy.