Conversion of CT image quality to suitable quality for images input into a computer-aided detection (CAD) system for lung cancer

当计算机辅助检测（CAD）应用于肺癌筛查时，选择适合于输入到CAD系统的图像的重构核是很重要的。然而，如果适用于CAD的核与放射科医生在引入CAD之前临床使用的核不同，核的改变可能会造成困难。为了克服这个问题，我们提出了一种新的图像滤波技术，作为应用不同重建核的替代方案，基于对系统中测量的调制传递函数（MTF）的定量分析。结果表明，该方法具有良好的性能，并改善了由于使用不适合图像输入的核而导致CAD系统性能下降的问题。

When computer-aided detection (CAD) is applied to lung cancer screening, it is important to choose a reconstruction kernel suitable for images input into a CAD system. However, if the kernel suitable for CAD is different from the kernel that has been clinically used by a radiologist until the introduction of CAD, the change in kernel could cause difficulty. To overcome this problem, we proposed a new technique of image filtering as an alternative to the application of a different reconstruction kernel, based on the quantitative analysis of the modulation transfer function (MTF) measured in the system. It was demonstrated that the proposed method had good performance and improved the worsening in performance of CAD caused by the use of kernels unsuitable for images input into a CAD system.