Evaluation of attenuation and scatter correction requirements as a function of object size in PET small animal imaging

In human emission tomography, an additional transmission scan is often required to obtain accurate attenuation maps for attenuation correction (AC) and scatter correction (SC). These methods have been translated to small animal imaging, although the impact of photon interactions on the reconstruction of mouse/rat images is substantially less than that in human imaging. In this study, we evaluate the requirement of these corrections for PET small animal imaging through Monte Carlo simulations of the Inveon PET scanner using MOBY voxelized phantoms scaled to different sizes (2.1-6.4 cm diameters). The 3D data were reconstructed in 6 different conditions depending on the attenuation map used: Accurate AC+SC, Simple AC+SC, Accurate AC only, Simple AC only, SC only and No correction. Mean error% for 8 different ROIs and 6 different MOBY sizes were obtained against the accurate reconstruction (first on the list). In addition, real mouse data obtained from an Inveon PET scanner were analyzed using similar methods. The results from simulations and real mouse data showed that attenuation correction based on solely emission data should be sufficient for imaging animals smaller than 4 cm diameter.