MTF Evaluation and Clinical Application according to the Characteristic Kernels in the Computed Tomography

Yoo, Beong Gyu; Kweon, Dae Cheol; Lee, Jong Seok

Abstract

Our objective was to evaluate the clinical feasibility of spatial domain filtering as an alternative to additional image reconstruction using different kernels in CT. Kernels were grouped as H30 (head medium smooth), B30 (body medium smooth), S80 (special) and U95 (ultra sharp). Derived from thin collimated source images, four sets of images were generated using phantom kernels. MTF (50%, 10%, 2%) measured with H30 (3.25, 5.68, 7.45 lp/cm), B30 (3.84, 6.25, 7.72 lp/cm), S80 (4.69, 9.49, 12.34 lp/cm), and U95 (14.19, 20.31, 24.67 lp/cm). Spatial resolution for the U95 kernel (0.6 mm) was 33.3% greater than that of the H30 and B30 (0.8 mm) kernels. Initially scanned kernels images were rated for subjective image quality, using a five-point scale. Image scanned with a convolution kernel led to an increase in noise (U95), whereas the results for CT attenuation coefficient were comparable. CT images increase the diagnostic accuracy in head (H30), abdomen (B30), temporal bone and lung (U95) kernels may be controlled by adjusting CT various algorithms, which should be adjusted to take into account the kernels of the CT undergoing the examination.