Implementation of an effective KL domain penalized weighted leastsquares sinogram restoration for low-dose CT colonography

Currently available spiral/helical computed tomography (HCT) technologies have demonstrated the potential for CTbased virtual colonoscopy or CT-colonography (CTC). However, a major limitation for this clinical application is associated with the risk of high radiation exposure, especially for its use for screening purpose at a large population. In this work, we presented an improved Karhunen-Loeve (KL) domain penalized weighted least-squares (PWLS) strategy which considers the data correlation among the nearby angular views and nearby axial slices simultaneously so that a fully three-dimensional (3D) restoration problem can be reduced to a 1D operation and therefore an analytical calculation is feasible for highest computing efficiency. The KL-PWLS strategy was implemented and tested on computer simulated sinograms which mimic low-dose CT scans at different noise levels characterized by 50, 40, 30, 20 and 10 mAs respectively. The reconstructed images by the presented strategy demonstrated the potential of ultra low-dose CT at as low as 10 mAs for VC application. Further evaluation by receiver operating characteristic study is needed and is under progress.