Effect of cone-beam CT detector size and projection geometry on scatter-induced image noise

Abstract

<h3>Objective</h3> The aim of this study was to evaluate the effect of detector size with different projection geometries on scatter-induced image noise in cone beam computed tomography (CBCT). <h3>Study Design</h3> A polymethyl methacrylate phantom containing air and aluminum underwent CBCT scanning positioned in the center of the FOV. Images were acquired with the PaX-i3D Green Premium (Vatech, Korea), which has a narrow vertical detector with off-axis projection geometry and multiple consecutive rotations for small and medium FOVs; the OP300 3D Maxio (Instrumentarium Dental, Tuusula, Finland), with aligned projection geometry for small FOV, and off-axis geometry for medium FOV; and the Promax 3D Max (Planmeca, Helsinki, Finland), which has aligned geometry for small and medium FOVs. Scans were performed using small and medium FOVs, combining tube voltages and tube currents to maintain a fixed normalized dose-area product. Signal difference-to-noise ratio (SDNR) was calculated on two-dimensional projection images and compared using one-way ANOVA with post hoc Tukey tests (α =0.05) between CBCT units. <h3>Results</h3> For small FOVs, Promax showed the highest SDNR compared to OP300 and PaX-i3D, with the difference being statistically significant (p < 0.0001). OP300 exhibited significantly higher SDNR values than PaX-i3D (p < 0.0001). For medium FOVs, Promax had significantly higher SDNR than OP300 and PaX-i3D (p < 0.0001), and no difference was discovered between OP300 and PaX-i3D. Therefore, Promax had the lowest scatter-induced image noise since it has aligned projection geometry in small and medium FOV sizes. For OP300, which has an off-axis geometry, the scatter-induced image noise with the medium FOV was comparable to PaX-i3D. <h3>Conclusion</h3> Reducing the active area of the detector does not result in fewer scattered photons reaching the detector, as it needs more rotations to capture all the projections. Moreover, in off-axis projections there is a lateral off-set of the detector, which results in the center of the FOV being positioned in the edges of the projection images. <b>Statement of Ethical Review</b> Ethical review was sought and study was exempted from ethical review