Introduction: Cone-beam computed tomography (CBCT) data are, in principle, metrically exact. However, clinicians need to consider the precision of measurements of dental morphology as well as other hard tissue structures. CBCT spatial resolution, and thus image reconstruction quality, is restricted by the acquisition voxel size. The aim of this study was to assess geometric discrepancies among 3-dimensional CBCT reconstructions relative to the micro-CT reference.
Methods: A total of 37 permanent teeth from 9 mandibles were scanned with CBCT 9500 and 9000 3D and micro-CT. After semiautomatic segmentation, reconstructions were obtained from CBCT acquisitions (voxel sizes 76, 200, and 300 μm) and from micro-CT (voxel size 41 μm). All reconstructions were positioned in the same plane by image registration. The topography of the geometric discrepancies was displayed by using a color map allowing the maximum differences to be located.
Results: The maximum differences were mainly found at the cervical margins and on the cusp tips or incisal edges. Geometric reconstruction discrepancies were significant at 300-μm resolution (P = .01, Wilcoxon test).
Conclusions: To study hard tissue morphology, CBCT acquisitions require voxel sizes smaller than 300 μm. This experimental study will have to be complemented by studies in vivo that consider the conditions of clinical practice.
Keywords: Cone-beam computed tomography; reconstruction quality; voxel size; x-ray micro tomography.
Copyright © 2014 American Association of Endodontists. Published by Elsevier Inc. All rights reserved.