Assessment of the prompt gamma coincidence correction approach using I-124: a phantom study

Ziel/Aim I-124 PET quantification is impaired because of the presence of prompt gamma coincidences (PGCs). The objective of the study was to assess the PGC correction approach using phantom setups under challenging conditions that are clinically observed in antibody and somatostatin receptor imaging.

Methodik/Methods All measurements were performed on a Siemens Biograph mCT PET/CT system with I-124 (and F-18 for reference purposes). Three types of phantoms were used: (a) a soft-tissue tumor phantom (consisting of spheres of varying sizes mounted in an abdominal phantom) that mimics tumors embedded in a warm background, (b) a bone-lung tumor phantom that simulates tumors in cold lung and cortical bone tissues, where the activity was filled only in spheres and in the abdominal cavity but not in the lung and cortical bone compartments, and (c) an organ phantom that simulates organs in a warm background with activity filled in spine, kidney and liver inserts as well as the abdominal cavity. A standard clinical acquisition protocol and a reference acquisition protocol was performed using an emission time of 240 s and 1 h, respectively. All I-124 data were reconstructed with (PGCon) and without PGC correction (PGCoff). The percentage differences between imaged and activity meter-based activity concentrations (ACs) were determined for PGCoff and PGCon and F-18 images.

Ergebnisse/Results For all phantom setups, PGCon images quantification were comparable to that of F-18 images (<10 %). AC deviations between PGCoff and PGCon (DPGC) images are similar for both acquisition protocols. For the phantom (a), the sphere DPGC (within parentheses the background) were, on average, −14 % (−30 %). For the phantom (b), sphere DPGC in cortical bone inserts (within parentheses in the lung insert) ranged from −90 % to −60 % (−1 % to −4 %). For the phantom (c), DPGC for the spine, kidney, and liver inserts were about –15 %, –10 %, and –25 %, respectively. The underestimation of the ACs in PGCoff images is a consequence of the scatter overcorrection in the single scatter simulation algorithm.

Schlussfolgerungen/Conclusions PGC correction approach is mandatory for reconstructing I-124 images to obtain higher quantitative accuracy, in particular for bone tumors and extended organs.