[Accuracy and reliability of three-dimensional craniofacial cone-beam CT superimposition method based on voxel registration].

[Accuracy and reliability of three-dimensional craniofacial cone-beam CT superimposition method based on voxel registration].

Zhonghua Kou Qiang Yi Xue Za Zhi. 2016 Aug 9;51(8):475-9

Authors: Zhou GH, Chen L, Sun YY, Ge N, Su L, Bai YX

Abstract
OBJECTIVE: To set up a three-dimensional(3D)craniofacial cone-beam CT(CBCT)superimposition method based on voxel registration, and evaluate the accuracy and reliability of different reference superimposition areas.
METHODS: CBCT scans were taken on a dry skull for three times with a time-interval of two weeks, and CBCT scans were taken on ten adult volunteers twice with a time-interval of one month. The 3D superimposition based on voxel registration was set up by means of Analyze 12.0. The bilateral-zygomatic bone and anterior cranial base were selected as different reference areas for voxel overlapping. The 3D overlapping images were output into Geomagic Qualify 2013, and the surface distances on different craniofacial regions were measured. Three operators repeated three times of these superimposition and measurement. SPSS 19.0 were used to analyze the accuracy and reliability.
RESULTS: When bilateral-zygomatic bone used as superimposition reference, the mean surface distance range of dry skull were from 0.035 to 0.064 mm, and volunteers from 0.099 to 0.182 mm. When the anterior cranial base used as superimposition reference, the mean surface distance range of dry skull were from 0.038 to 0.071 mm, and volunteers from 0.127 to 0.218 mm. All these mean values were less than the CBCT voxel size 0.25 mm(P<0.05), and got sub-pixel precision. No significant difference was found between the different operators at different times(P>0.05).
CONCLUSIONS: This study showed the high accuracy and reliability of 3D CBCT superimposition based on voxel registration, either the anterior cranial base or the bilateral-zygomatic bone as reference. This method could be used for evaluating 3D craniofacial treatment effects.

PMID: 27511038 [PubMed - in process]

from #ENT-PubMed via ola Kala on Inoreader http://ift.tt/2bcbJGu
via IFTTT