Evaluation of Cone-beam Computed Tomography Artifacts Produced by Metal Objects Located Within and Outside the Field of View.
Poorya Jalali, D.D.S.
Kassabji, A., Tahmasbi, M., Augsburger, R.A., Nair, M., Kesterke, M.J. and Jalali, P. (2021). “Evaluation of Cone-beam Computed Tomography Artifacts Produced by Metal Objects Located Within and Outside the Field of View.” J Endod Dec 8;S0099-2399(21)00869-4. [Epub ahead of print].
INTRODUCTION: Artifacts created by the presence of metal objects in the jaw negatively affect the cone-beam computed tomographic image quality. This study compares artifacts produced by metal objects outside of the field of view (FOV) in a small FOV CBCT image with those produced in a large FOV image in which the metal object is within the FOV. METHODS: We methodically placed 4 titanium implant-sized rods and 4 zirconium crown-sized disks on 1 side of a human cadaver mandible. Using the Accuitomo 170 CBCT machine (J Morita, Irvine, CA), a total of 18 scans (9 with a small FOV and 9 with a large FOV) were made. Ten axial slices from each scan were transferred to ImageJ software (National Institutes of Health, Bethesda, MD) for analysis. The mean standard deviation of all voxel values of a fixed region of interest (ie, uniform air located lingual to tooth #30) was compared between small and large FOV slices. Two blinded observers subjectively rated the images for diagnostic quality and the presence of artifacts. RESULTS: The Wilcoxon signed rank test showed that the standard deviation for both small and large FOV slices increases as the number of metal objects increases. The mean of the standard deviation for small and large FOVs is 3.6 and 2.5, respectively (P = .0000). Fifty-three percent of the small FOV slices had more artifacts in the subjective analysis. One hundred percent of the small FOV slices were rated as higher quality. CONCLUSION: Metal objects outside of the FOV in the contralateral quadrant do affect the quality of small FOV images. However, small FOV images have a higher resolution compared with large FOV images.