Post-reconstruction-based partial volume correction methods: A comprehensive review

The vital role of positron emission tomography (PET) image is to provide an accurate quantitative information. Nevertheless, the partial volume effect (PVE) can easily alter the quantification. This effect is a consequence of the limited spatial resolution in PET which is due to the non-collinear and/or inherently random photon. For several years, great effort has been devoted to study the partial volume correction (PVC) and improve the quality of the image. At first stage, only PET modality was used and then, further study of the issue involved anatomic information from high resolution modalities in the correction process. Clearly, this regularization using segmentation, correlation and the registration step between the two modalities enhanced the performance of these techniques. These methods can be, mostly, divided into two main approaches: reconstruction-based methods and a post-reconstruction-based methods. In this paper the focus of attention was to present the most used post-reconstruction PVC methods developed in the literature. We will introduce the principal of each method, its extension and their applications in different domains.