The current scanning parameters for computed tomography (CT) such as multi-detector row CT are becoming more complicated, and there are many cases in which the selected parameters directly affect image quality. Therefore, to evaluate the effectiveness and validity of the selected parameter, quantitative image quality evaluations are indispensable. Among the items of evaluation, modulation transfer function (MTF) is one of the most important in evaluating the resolution property. Several guidelines for performance evaluation for CT have been reported since the era of early CT diffusion. In those guidelines, it is recommended that the resolution property needs to be measured by the wire method, in which a phantom designed to support a thin metal wire along an axis perpendicular to the slice plane is used. However, the academic papers describing the wire method are so old that the detailed methodology corresponding to currently available CT specifications cannot be conducted. However, the fundamental principles are still effective. In this study, we examined the calculation method, phantom design and allocation, wire material, and other factors suitable for current CT specifications, and derived some recommendations from them.
In three-dimensional CT angiography (3D-CTA), good reproducibility can be obtained by maintaining the maximum CT numbers (HU) at a specified level. However, the correlation between the scan time and the injection time showed that the maximum CT numbers increased and varied due to the additional contrast enhancement effect from recirculation of the injected contrast medium for longer injection times when the dose of iodinated contrast medium per unit time (mgI/s) was maintained at a specified level based on the time-density curve (TDC) of the phantom. The amount of contrast medium employed at our hospital has been optimized based on an iodinated contrast medium dose per unit time providing a contrast enhancement effect of 300 HU in the middle cerebral artery. Using this standard, a TDC phantom was employed to obtain an iodinated contrast medium dose per unit time, permitting equivalent maximum CT values (used as standard values) to be obtained by changing the injection time. A contrast-enhancement technique that accounts for the variation in the scan time was evaluated. Strong correlations were observed between the scan time and the injection time (R2=0.969) and between the injection time and the dose of iodinated contrast medium per unit body weight (R2=0.994). We conclude that adjusting the dose of iodinated contrast medium per unit body weight per unit time according to the scan time permits optimization of the contrast-enhancement technique.
Window setting is a very important technique in CT examinations. However, most beginner technologists have difficulty in setting the optimal window. Now, thanks to technical progress, it is easy to obtain a great many CT images. On the other hand, it is impossible to provide the optimal window setting for all images. Therefore, our purpose is to offer optimal CT images for every patient by using the automatic window-level and width-setting system. As a result of this experiment, there was a considerable difference in window setting by an expert technologist and that by a beginner technologist. With our system, we were able always to obtain an optimal window setting, such as that set by an expert technologist, regardless of the CT experience of the radiological technologist. We think that this system will be effective in observing animated examinations even if film is no longer used in the future.
Few studies have investigated the detailed imaging characteristics of multiplanar reconstruction (MPR) images, which have come to be used in imaging diagnosis. The purpose of this study was to evaluate the slice profile characteristics of MPR images. The slice profile of a coronal image was measured by the bead method. Moreover, it assumed that the slice profile of an MPR image became a convolution of the square profile corresponding to the nominal slice thickness and line spread function (LSF) of an axial image, and the simulation was performed. The nominal slice thicknesses of the original axial image and coronal image were 1.0 mm, 2.0 mm, and 3.0 mm. Three reconstruction kernels (B20, B30, and B40) of the original axial image were used. The results of measurement revealed that the full width at half maximum (FWHM) values were 1.7 mm for reconstruction kernel B20 and 1.3 mm for reconstruction kernel B40 in the case of a nominal slice thickness of 1.0 mm. The simulated and measured modulation transfer factors (MTF) were in close agreement. Then the slice profile of the coronal (sagittal) MPR image forms by the convolution of a LSF of the y- (x-) direction and the square profile with a nominal MPR slice width, and is affected by the reconstruction kernel.
In chest CT images, the dorsal lower lung field often shows an infiltration-like shadow in patients who cannot stop breathing or take a deep breath. The cause of this phenomenon might be due to the effects of gravity. Since we had observed decreased effects of gravity by conducting additional CT scanning for patients in an oblique position (55°) or a nearly lateral position, we conducted a clinical study to investigate this matter. Forty-three patients (23 patients in the normal group and 20 patients in the inflammatory disease group) who underwent additional CT scanning were included in this study. CT values for the region in which infiltration-like shadow was observed in both positions (dorsal position and oblique position) were measured. The ratio of fluctuation in the CT value of the dorsal lower lung field at a positional change from the dorsal to the oblique position was calculated as a coefficient of fluctuation C (%). As a result, the coefficient of fluctuation C (%) was 32.6±13.6 in the normal group and 6.7±6.8 in the inflammatory disease group. The effects of gravity were improved by additional CT scanning in an oblique position (55°) or a nearly lateral position, and this enabled differentiation of the effects of gravity vs. inflammatory diseases.
The use of CT as a general examination has spread widely and is even used in small institutions. However, it is difficult to determine the current situation of each institution. Therefore, we employed a questionnaire to investigate the current situation of a variety of institutions. From the results of the questionnaire, we determined that the window setting was difficult for beginner technologists. In addition, in many institutions, radiological technologists did not always use the same display FOV for the same patient. From this questionnaire, we were able to determine the present conditions in each institution. We consider these results very useful.