Comparison of Cone Beam Computed Tomography Scans With and Without Simulation of Head Motion

Purpose: To test for differences in image quality between 3D volumetric datasets acquired by cone beam computed tomography (CBCT) with and without simulation of head motion by visual analysis of individual image sections and by comparison of scan-based distance and Hounsfield unit (HU) measurements. Materials and Methods: A total of 200 volumetric datasets were acquired with and without remote-controlled, simulated movement of a human cadaver head using the KaVo 3D eXam CBCT system. Results: The "Landscape 8 x 8 cm Slide" mode provided a sufficient field of view at a low radiation dose. All datasets showed reproducible results. Our analysis showed that the level of image quality and image detail increased with increasing resolution. Linear distance measurements and HU measurements in the cone beam CT scans acquired without simulation of head motion were absolutely comparable to those obtained with head motion simulation. Conclusions: There is no practical difference in image quality between CBCT scans acquired with and without head motion if no long-term change in head position occurs during the acquisition process. Slight head movement has no clinically relevant effect on the geometric accuracy or visual image quality of cone beam CT scans. Keywords: Cone beam computed tomography, motion artifact, head motion, distance measurement, Hounsfield unit, image quality