Accuracy Assessment of Linear Measurements Obtained by Cone Beam Computed Tomography at Different Voxel Resolutions: Comparative in-vitro Study

Abstract

Purpose: To determine the accuracy of linear measurements (LM) at different voxel resolutions on various anatomic locations in cone beam computed tomography (CBCT).
Materials and Method: A total of 25 dried mandibles were used & 75 scans were performed in this study. 12 radiopaque markers were placed in the holes created by round carbide bur at different positions leading to 6 readings which were horizontal & longitudinal measurements in each mandible at various sites. The LM readings were taken first with the digital vernier calliper considering it as gold standard and later on CBCT at three different voxel resolutions i.e. 0.15mm, 0.30mm and 0.45mm by two observers blindly. FOV of the scan was 8 x 5 cm. The measurements were made between nearest points of radiopaque markers. ANOVA, post hoc test and intra class correlation was applied for comparison of the variables. P value < 0.05 was considered significant.
Results: On comparing the measurements taken by vernier calliper and at different voxel sizes of CBCT, no statistical significant different was found. Also, the post hoc analysis revealed no significant difference between measurements done by vernier calliper and CBCT at different voxel resolutions.
Conclusion: The measurements made on CBCT are strongly reliable even on changing the voxel size making it an efficient imaging tool in dentistry. LM made at higher voxel resolutions reduces patient exposure and are equally reliable.