PET Image Reconstruction Based on a Constrained Filter

Abstract The paper aims to propose an algorithm for PET(positron emission tomography) image reconstruction. Different from the algorithms designed to satisfy a single performance index in most cases, our algorithm inherently involves for two objects. Due to that the information upon noises is utilized fully, our algorithm provides a better reconstructed results than the one for the a single index when one obtains only some information on noises but does not know whether the statistics matches the noise level well in advance. The conclusion is also supported by the numerical experiments.