Clinical NECR in 18F-FDG PET scans: Optimization of patient specific activity and variable acquisition time. Relationship with SNR

It is common to optimize the patient specific activity and the acquisition time per bed position for 18F-FDG PET scans by using metrics extracted from phantom experiments. However, optimal activity and time duration can significantly vary from experimental set-up and from patient to patient. An approach using a patient-specific noise equivalent count rate (NECR) has been proposed and showed that it could be used for optimizing clinical scanning protocols for 18F-FDG-based acquisitions. In this work, we propose to use the clinical NECR on a large population as a function of the body mass index (BMI) for deriving the optimal injected activity (IA) and the optimal acquisition time per bed position (oAT). The relationship between the NECR and the signal-to-noise ratio (SNR) was also assessed both in an experimental and a clinical setting. The Siemens Biograph mCT TOF scanner with an extended field-of-view (FOV) was used. 491 consecutive patients were retrospectively evaluated and divided into 4 BMI subgroups for the IA and oAT studies. The NECR was modeled for each patient taking into account the urinary excretion. Two dosing regimens were assessed by computation of activity that corresponded to 90% of the peak NECR and that reached a pre-defined NEC value. The oAT was adjusted by weighting the time per bed position by the NECR value while keeping the total acquisition time constant. Finally, the relationship between NECR and SNR was investigated using an anthropomorphic phantom scanned 36 times over 12h for the experimental part and based on a population of 507 other patients for the clinical evaluation. While the first dose regimen suggested a unique IA (665 MBq) regardless of the BMI, the second dose regimen proposed a variable IA and a total acquisition time according to the BMI with a maximum activity for each class of BMI. The NEC improvement was 35% as compared with the local current injection rule. Clinical NECR modeling made it possible to derive variable time- per bed position according to the BMI and anatomical region. NEC and number of true events were found to be highly correlated with SNR for the experimental set-up. This result was moderately confirmed in the patient study regardless of the BMI subgroup suggesting that for the scanner and nonlinear reconstruction algorithm used in this study, NEC or the number of true events linearly correlated with SNR. Finally, our work showed that clinical NECR modeling was a valuable alternative to phantom-based experiments