A simulation study for optimizing the injected dose of clinical PET systems

The optimization of the injected dose in PET imaging systems is important for the design of clinical data acquisition protocols. Methods to reduce the total amount of radioactive dose injected into a patient are investigated. On the other hand, a lower dose may require a longer data acquisition time to obtain images of high statistical quality, thus limiting the total number of PET scans performed by a PET facility per day. Through dose optimization, a compromise between the total dose injected into a patient and the required scan time can be achieved, by ensuring maximum count rate performance for the scanner. In this study, we use the Noise Equivalent Count Rate (NECR) as a metric of the rate in which statistically important coincidence events are counted by a PET imaging system. The optimal dose is defined as the total amount of dose required to be injected into a patient so as to induce a maximum NECR for the particular scanner and patient. Ideally, it can be estimated by determining a curve of NECR versus the injected dose for each scanner-patient system. Recent studies estimate optimal dose for a particular system by expressing the NECR as a function of the singles rate instead. In this work we propose an alternative method based on a series of simulations of imaging systems and realistic anthropomorphic phantoms. We used Geant4 Application for Tomography Emission (GATE) to simulate a series of scans and study the independent effect of various parameters to the NECR of a clinical PET system. We investigated the relationship between the NECR and the parameters of patient size, relative axial position of the patient to the field of view (FOV) of the scanner, combined use of reduced dead time electronics and LSO crystals instead of slow-responding electronics and BGO and finally of the energy window of a system. We used two validated scanner models (ECAT HR+ and Biograph), three NCAT phantoms of different size, two bed positions corresponding to the axial shift of either the heart or the bladder region to the centre of FOV respectively and three energy windows with different low energy thresholds. The results show an optimal mean dose range of 55-65MBq for HR+ and 300-450MBq for Biograph, depending on the patient size, when the heart is located at the centre of axial FOV.