Iterative image reconstruction for sparse-view CT using normal-dose image induced total variation prior

Sparse-view x-ray computed tomography (CT) imaging still is an interesting topic in CT field. In this paper, a new iterative image reconstruction approach for sparse-view CT with a normal-dose image was presented. The proposed cost-function which is under the criteria of penalized weighed least-square (PWLS) for CT image reconstruction mainly contains two terms, i.e., fidelity term and prior term. For the fidelity term, the weights of weighed least-square term are determined by considering the relationship between the variance and mean of the projection data in the presence of electronic background noise. For the prior term, a normal-dose image induced total variation (ndiTV) prior is proposed as an extension of the PICCS algorithm introduced by Chen et al 2008, which can relieve the requirement of misalignment reduction of the PICCS algorithm. For simplicity, the present approach is referred to as "PWLS-ndiTV". Qualitative and quantitative evaluations were carried out on the present PWLS-ndiTV approach. Experimental results show that the present PWLS-ndiTV approach can achieve significant gains than the existing similar methods in noise and artifacts suppression.