Investigation of the minimum detectable activity level of a preclinical LSO PET scanner

Novel molecular imaging applications increasingly involve studies where very low amount of activity is present. This operation point can be challenging in terms of image quality especially when the intrinsic detector activity from scintillators such as LSO is considered. LSO crystals contain ¹⁷⁶Lu which emits beta^- particles followed by gamma photons, resulting in the detection of true and random events. This background activity has been shown to contribute a significant percentage of the total detected true events, when a very weak activity distribution is imaged. This can affect the weakest signal that can be detected by an LSO PET scanner which determines its "detection limit" and is evaluated by the parameter of minimum detectable activity or MDA. A series of acquisitions was performed in order to study the effect of the energy window to the intrinsic true and randoms rate. The experiments were also simulated with GATE, and the results were validated by comparing them with the experiment. Four square regions each with a unique signal to background activity concentration ratio were used. The background activity level was held constant between the different regions, while the activity of the point sources varied in 4 selected levels. The signal to background ratio was calculated separately for each region. The energy spectrum of the intrinsic background activity and its contribution to the total energy spectrum both for singles and coincidences was estimated through the GATE simulation. We histogrammed both the measured and simulated data on various time frames, which we reconstructed using the filtered backprojection algorithm. Every image was quantified based on the Currie equation in order to associate an MDA value for each of the 4 point sources as a function of the frame length. In the case of the microPET Focus 220 a total amount of 4nCi/mm³ can be reliably detected for frame lengths longer than 5 min and at regions where the- signal to background activity concentration ratio is higher than 4. In the case of higher contrast regions detection can be achieved even for frame lengths down to 1 min.