Purpose: The aim of the study was to find out whether in [F]fluoro-2-deoxy-D-glucose positron emission tomography (F-18 FDG-PET) performed in children in our department, the number of counts were sufficiently high to guarantee optimal image quality.
Methods: Phantom data with negative and positive lesions were acquired with an increasing count density and the relationship between image quality and counts per voxel was analyzed. On PET images obtained in 28 children aged 4-17 years and using a standard administered activity of 6.5 MBq/kg with a minimum of 74 MBq, the maximum counts per voxel in 10 regions of interest were noted and compared with those obtained on phantoms.
Results: Phantom data showed clearly the effect of higher count on image quality. Moreover, when high counts were available, reconstruction could be done using 2 mm voxel size, which improved the quality of the images further. In children, the counts per voxel in all the regions of interest, with the exception of the brain, were at a level of suboptimal image quality.
Conclusion: By following the official guidelines, the counts per voxel observed in children with our gadolinium oxyorthosilicate-PET system were lower than the minimal required to produce optimal quality images. More efficient PET systems or higher F-18 FDG dose or increasing the acquisition time or combination of these factors are needed to improve the image quality and lesion detection.