Low Dose CT Filtering in the Image Domain Using MAP Algorithms

The purpose of this paper is to present two new noise reduction filters techniques in the CT image space, in order to provide a better quality to the images acquired with low radiation exposure. For the noise reduction, a new denoising technique is presented based on a pointwise Maximum a Posteriori (MAP). The noise is considered Gaussian with zero mean, as observed experimentally, and the variance is estimated considering a signal-independent noise. For the a priori density of the signal, we used different non-negative probability densities (reflecting the fact that the pixels of an image are non-negative). In another approach, the histogram of the images were segmented into unimodal parts and each segment was filtered using the filter based on the MAP criterion with the a priori density that best fits it. After filtering, the evaluation of the method is performed using the following criteria: Peak Signal-to-Noise Ratio, Universal Image Quality Index and Structural Similarity Index. The 2D filtering results are compared with the results obtained by pointwise Wiener filter. Simulation and real CT images results show that the proposed techniques increase the image quality and improve the use of a low-dose CT protocol.