An effect of time-of-flight (TOF) information on normalization components and point spread function (PSF) in positron emission tomography (PET)

Abstract

Before the image reconstruction process takes place, emission data from a PET scanner must be corrected for scatter and changes for detector sensitivity. Normalization process should take care of the variations in the detector sensitivity. Because of the need of high statistics for large number of line-of-responses (LORs), component-based normalization method was implemented in the state-of-the-art PET scanners. Since the development of TOF PET scanners, normalization components were never revisited to examine the effect of the TOF dependency. The purpose of this work was to understand the TOF dependency on normalization. Experiments and Monte Carlo simulations were acquired to examine the TOF effect on normalization. In this process, point spread function was also investigated to understand the TOF dependency through the depth of the LOR. PSF responses at four locations have been estimated through Monte Carlo simulations along two LORs at two positions each. Preliminary PSF measurements were also performed by using a PET scanner for comparison.