A Comparison of Filtered Back Projection and Maximum Likelihood Expected Maximization

Abstract

In this study, we compare two commonly used image reconstruction methods which are the filtered back projection (FBP) and the maximum likelihood expectation maximization (ML-EM) on some medical and phantom image with noise. To evaluate those methods, we used one evaluation measurement which is called a peak signal-to-noise ratio. It is most commonly used to measure the quality of reconstruction. In this experiment, the methods are tested with two images of computer tomography, two phantom images, and one SPECT images. Experimental result shows that FBP and ML-EM are closely similar result but MLEM is better than FBP in noisy images.