Using Monte-Carlo simulations to implement corrections for I-124 as a non-pure positron emitter in small animal and human PET imaging

Using I-124 for PET imaging applications implies some difficulties concerning the image quality: The resolution is degraded by the large maximum positron energy of 2.1 MeV and the resultant long positron range. In addition I-124 is a non-pure PET isotope exhibiting additional gamma ray emissions with high contributions to the total decay scheme: 602.72 keV with 63% and 722.78 keV with 10%. These fractions cannot be quantified exactly in PET measurements. Therefore in our work we utilized GATE 6.1 [1] to investigate the effects of these supplementary "false" coincidences on a spectral, sinogram- and image-based basis. Three PET systems were modeled with GATE and confirming measurements were accomplished on them: Two small animal PET scanners (Raytest ClearPET and Concorde MicroPET Focus 120) and one human scanner (Philips Gemini TF 64). Derived from the simulated energy spectra, we propose narrower energy window configurations depending on the energy resolution of the different systems in order to minimize the amount of false coincidences. Separating the simulated sinograms for true and false coincidences revealed that a two-component correction for I-124 has to be implemented. A homogenous background subtraction is amended by a Gaussian-shaped fit with factors taken from GATE simulations which considers an additional portion of false coincidences within the phantom borders. Both methods, the energy windowing and the sinogram-based background correction, were successfully applied to measured data for different phantom geometries for the MicroPET scanner. Nevertheless comparative simulations of the MOBY mouse voxel phantom [2] with F-18 showed that the predominant effect in small animal imaging is the resolution loss due to the high positron range.