Noise reduction for low-dose single-slice Helical CT sinogram

Abstract

Helical computed tomography (HCT) offers several advantages on conventional step-and-shoot CT for imaging a relatively large object, especially in dynamic studies. However, it may increase the X-ray exposure significantly. This work aims to reduce the radiation by noise reduction on low-dose (or mA) sinogram of HCT. The noise reduction method is based on three observations on HCT: (1) the axial sampling of HCT projections is nearly continuous as the detector system rotates; (2) the noise distribution in the sinogram space is nearly a Gaussian after system calibration (including logarithmic transform); and (3) the relationship of calibrated data mean and variance can be expressed as an exponential functional across the field-of-view. Based on the second and third observations, a penalized weighted least-square (PWLS) solution was chosen, where the weight is given by the data mean-variance relationship. The first observation encourages the use of Karhunen-Loeve (KL) strategy along the axial direction. In the KL domain, the eigenvalue of each principal component was used for an adaptive noise smoothing via the penalty. The KL-PWLS noise-reduction method was implemented analytically for efficient reconstruction of large volume HCT images. Simulation studies demonstrated noticeable improvement, in terms of image quality measures and abnormal detectability observer studies, of the proposed noise-reduction method over conventional low-pass noise filtering with an optimal cutoff frequency and/or other filter parameters