Modulation transfer function evaluation of cone beam computed tomography for dental use with the oversampling method

Abstract

Objectives: The aim was to investigate the possibility of evaluating the modulation transfer function (MTF) of cone beam CT (CBCT) for dental use using the oversampling method.

Methods: The CBCT apparatus (3D Accuitomo) with an image intensifier was used with a 100 mum tungsten wire placed inside the scanner at a slight angle to the plane perpendicular to the plane of interest and scanned. 200 contiguous reconstructed images were used to obtain the oversampling line-spread function (LSF). The MTF curve was obtained by computing the Fourier transformation from the oversampled LSF. Line pair tests were also performed using Catphan(R).

Results: The oversampling method provided smooth and reproducible MTF curves. The MTF curves revealed that the spatial resolution in the z-axis direction was significantly higher than that in the axial direction. This result was also confirmed by the line pair test.

Conclusions: MTF analysis was performed successfully using the oversampling method. In addition, this study clarified that the 3D Accuitomo had high spatial resolution, especially in the z-axis direction.

Figure 1

Reconstructed images of the 100 μm tungsten wire. (a and b) Images in the axial and z-axis directions, respectively. The white squares surrounding the wire represent the region of interest, the size of which is 30×30 pixels. For the modulation transfer function determination, the line-spread function was obtained by integrating in the direction illustrated by the arrows

Figure 2

The oversampling method and its result. (a) A relationship between slice number (S) and the centre of mass position of 200 contiguous images, which enabled the determination of the linear regression parameters. Based on these parameters, the 200 line-spread functions (LSFs) were rebinned to one-tenth of a pixel width (0.0125 mm), and the averaged LSFs obtained are shown in (b)

Figure 3

Modulation transfer function curves of cone beam CT in the axial and z-axis directions are illustrated. The data are represented as the mean and SD of five independent measurements

Figure 4

Axial (a) and z-axis (b) images of the high-resolution insert of Catphan®. The window width and window level were set to 3600 and 2400, respectively, in both images. (A), (B) and (C) represent 19, 20 and 21 lp cm⁻¹, respectively. In (b), 19–21 lp cm⁻¹ can be resolved, whereas, in (a), 21 lp cm⁻¹ cannot be resolved. These results are consistent with those by the analysis of the MTF curves shown in Figure 3