Comparison of Normalization Methods for ${^{124}}{\rm I}$ PET on Siemens Inveon PET Scanner

Abstract

<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">68</sup> Ge has conventionally been used for the normalization of <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">18</sup> F PET. However, because <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">124</sup> I PET has different characteristics such as a low positron branching ratio (23%), the presence of higher gamma energy (602 keV and 723 keV), and a larger positron range, the use of <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">68</sup> Ge should be validated for <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">124</sup> I PET. To the best of our knowledge, there has been no previous report on the effect of normalization between conventional <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">68</sup> Ge and <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">124</sup> I. In this study, we examined image quality to assess the effect of normalization. For this purpose, we measured the non-uniformities (NUs) and recovery coefficients (RCs) using a NEMA NU4 image quality phantom (NU4 phantom) and compared the values for the conventionally used <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">68</sup> Ge and <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">124</sup> I. A Siemens Inveon PET scanner was used throughout. Emission data were obtained using the NU4 phantom filled with <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">124</sup> I solution (3.7 MBq) for 20 min within 250-750 keV. The normalization data were acquired for 40 h using a cylindrical phantom (diameter: 6 cm, length: 13.5 cm) filled with <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">124</sup> I in a water solution (9.25 MBq). To match the energy window of an emission scan, the energy window setting for the normalization was 250-750 keV. The obtained list mode data were sorted into 3D sinograms. For comparison, normalization data were acquired using a cylindrical <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">68</sup> Ge normalization phantom (diameter: 6 cm, length: 15 cm). The data with no normalization correction were also acquired. The list mode emission PET data were reconstructed using FBP with a ramp filter and normalization, and attenuation and scatter correction were applied. The image quality of the reconstructed images was assessed in terms of the radial, axial, and volume NUs and RCs as a standard using the NEMA NU4 protocol. There were no significant differences in the NUs and RCs between <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">68</sup> Ge and <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">124</sup> I. These results indicate that conventional <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">68</sup> Ge can be used for <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">124</sup> I PET.