Detection of alveolar bone defects with three different voxel sizes of cone-beam computed tomography: an in vitro study.

This study was conducted to assess the accuracy of cone-beam computed tomography (CBCT) of different voxel sizes in the detection of alveolar bone defects, and to select the optimal voxel size for clinical use. 46 in-vitro teeth were placed in bovine ribs in which alveolar bone defects were randomly simulated. In total, 32 alveolar bone defects and 14 teeth without periodontal defects were used. CBCT images were acquired with the use of three different voxel sizes: 0.125-mm, 0.2-mm and 0.4-mm. The scan data were 3D-reconstructed in Mimics software and evaluated by two observers with more than 5 years of experience in CBCT. Receiver operating characteristic (ROC) curves and diagnostic values were obtained. Pairwise comparison of ROC curves was made for evaluation of the diagnostic values of different voxel sizes. Kappa statistics assessed the observer reliability. Results were considered significant at P < 0.05. It showed no statistically significant difference between 0.125-mm group and 0.2-mm group, but 0.4-mm group had lower Az values that differed significantly from 0.125-mm and 0.2-mm groups (P < 0.05). Based on diagnostic value and radiation protection, 0.2-mm voxel size may be a good choice for the detection of bone defects with CBCT.