In multislice CT, the detector elements are placed in arrays aligned along the z-axis. This geometry necessarily results in a greater angle of incidence for the X-ray beams at the outer rows of the detector array. This angle of X-ray beam incidence along the z-axis in a multislice detector is called the "cone angle "(see figure). The cone angle is thought to be a major cause of reconstruction error in multislice helical CT.
Image reconstruction of current multislice helical CT is based on the variation of the so-called "z-axis interpolation" method. This type of algorithm can be regarded as an extension of conventional helical interpolation algorithms, since the data of interest is generated by interpolation of neighboring data along z-axis. In z-axis interpolation, the data is back projected as a parallel beam regardless of cone angle. This can cause shading or streak-type artifact.
As long as z-axis interpolation is employed, it is not possible to totally eliminate the effect of cone angle; however, there are several methods for minimizing them. These are related to geometrical design of the scanner, the scanning technique, and the type of algorithm.
The focus-detector distance (FDD) is the most important factor. As is clearly shown in the figure, a shorter FDD inevitably results in a greater cone angle. Therefore, it is important for a multislice CT scanner to have a longer FDD in order to minimize the effect of cone angle.
In addition to a smaller cone angle, a longer FDD also brings with it the benefit of a longer focus-isocenter distance (FID) in the scanner, which is advantageous in reducing patient dose.
The table shows the FDD and FID of current multislice CT scanners (unit: mm)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
On the other hand, a shorter FDD allows a smaller X-ray tube to be used. Since X-ray attenuation is directly related to the square of the distance, a short FDD makes it easier to maintain sufficient output of X-ray photons even with a small X-ray tube. Another advantage of a shorter FDD is that the scanner can be made more compact. These two advantages of a shorter FDD are also related to the cost of the scanner.
I personally believe that the focus-detector distance of a particular scanner can serve as a useful index for determining "how true to the physics your scanner is."
Another factor that affects the cone angle is the width of the detector elements along z-axis. A wider detector aperture results in a larger cone angle, given the same FDD. This means that the use of a smaller slice thickness is an effective way to reduce these artifacts.
It should be noted that the FDD and detector aperture should always be considered together when we evaluate the effects of the cone angle. It is possible to employ a wider detector aperture when a longer FDD is used. In other words, it is possible to obtain better image quality with the same detector aperture when the FDD is longer.
The effects of the cone angle can be reduced by employing a specific sampling pitch and a special type of reconstruction algorithm (see MUSCOT).
