Bayesian sinogram smoothing with an anisotropic diffusion weighted prior for low-dose X-ray computed tomography

As is known, low-dose computed tomography (CT) image can be severely degraded by the excessive quantum noise. In order to address this problem, we firstly present a novel anisotropic diffusion weighted prior applied in Bayesian-based statistical sinogram smoothing approach in this work. Then, the reconstructed image is obtained by the filtered back-projection (FBP) from the smoothed projection data. Compared with the traditional priors, the proposed novel prior can adaptively adjust the smoothing degree according to the sinogram characteristic. The effectiveness and feasibility of the proposed approach are validated by both digital phantom and clinical data experiments. The superiority of the presented method over other methods is also quantitatively studied by resolution–noise tradeoff curves and signal to noise ratio (SNR). The experimental results indicate that the developed approach has the excellent performance for low-dose CT imaging.