Impact of time-of-flight and point-spread function modeling on the optimization of FDG PET/CT scan duration

2120 Objectives We investigated the effect of Time-of-Flight (TOF) algorithm and point spread function (PSF) based reconstruction on tumor detection and quantification in short time FDG PET/CT acquisitions. Methods Eighteen whole-body PET/CT examinations were performed on a Siemens Biograph mCT. Emission data were acquired 60 min after i.v. injection of 5 MBq/kg of FDG. A 3 min list-mode acquisition was resampled to simulate 180 s, 120 s, 90 s, 60 s, 30 s and 15 s acquisitions. PET images were reconstructed using AW-OSEM with TOF and PSF (3 iter, 21 subsets, 2 mm). Individual lesions were detected on the sets of images and semiquantitative analysis using volumes of interest was performed. The relative change in SUVmax and SUVmean of lesions with shorter scan times was reported. Furthermore, SUV values were measured for a 2 cm spherical region in the liver and SNR was determined. Statistical analysis was performed to evaluate differences across different time acquisitions. Results Visual analysis showed an increase in noise and a decrease in image quality as the acquisition time decreased. Sixty-six lesions were detected in 180 s, 120s, 90 s, and 60 s; 61 lesions were detected in 30 s and only 43 lesions were detected in 15 s. Decreasing the time per bed position to 120 s, 90 s, and 60s did not produced a substantial variation in SUV values. Scan times of 30 s and 15 s produced a significant increase in both SUVmax (5.8% and 21.2% respectively) and SUVmean (7.0 %, and 20.2% respectively). SNR progressively decreased as scan time decreased but there was not significant differences in SUVmean in liver tissue among the different times per bed position. Conclusions In whole-body FDG-PET/CT with TOF algorithm and PSF-based PET reconstruction the time per bed position affects SUVmax measurements. PET acquisition time can be reduced to 60 s/bed position with similar tumor detection and without significant variation in quantification, thus improving patient comfort.