Ultra-dense ceramic scintillator for BrainPET scanner

Summary Positron emission tomography (PET) is a powerful tool for imaging a wide range of biological processes in vivo and for detecting and monitoring human disease, as well as an important technique in neuroscience research. For human brain studies, PET offers a sensitive and noninvasive window into brain metabolism and neurochemistry, providing information from living human subjects that cannot be obtained using any other technique. The key characteristics of brain PET are spatial resolution and sensitivity. Brain PETs with a spatial resolution approaching 1 mm at the center of the field-of-view (FOV) are under development using LYSO or LSO scintillators, which are the currently available scintillators that can provide the best overall performance. However, to obtain a high sensitivity, which strongly affects the quantitative precision of PET imaging, L(Y)SO with a thickness of around 20 mm is widely used. The first and the major drawback of using the thick L(Y)SO crystal is the parallel error, which decreases the spatial resolution off the center of the FOV. Depth-of-Interaction (DOI) information can be used to improve spatial resolution off the center of the FOV, however, the best DOI resolution from 20 mm thick L(Y)SO is ~2.0 mm obtained using the dual-ended readout, which is not good enough. The second drawback of the L(Y)SO is the high cost. For most scanners, the L(Y)SO cost is half of the total scanner cost. The performance of brain PET scanners can be improved using a new scintillator based on Lu2O3 (LO), which has a higher density (9.4 vs. 7.4 g/cm3 for LSO) and higher effective Z (68 vs. 65 for LSO). When doped with La3+ and Yb3+, it exhibits an ultra-fast rise time (<0.6ns) combined with prompt charge transfer luminescence with decay time of 1.7 ns, giving timing resolution (TR) of 180 ps FWHM. The energy resolution at 511 keV is better than 10% (>10% for LYSO), due to the good light yield from slower components (<20,000 ph/MeV) and good proportionality (better than for LYSO). Hence, LO:La,Yb can be an excellent choice for a brain PET scanner requiring high sensitivity and high spatial resolution. The LO is produced using ceramic methods which may also have an edge over the crystal growth methods in terms of cost of manufacturing. Therefore, for this project, we propose to 1) optimize the LO:La,Yb and its manufacturing for high-resolution and high-sensitivity brain PET applications, 2) optimize LO:La,Yb based PET detector for brain PET through experimental evaluation, and 3) to estimate the performance of a full brain PET system based on these results through GATE Monte Carlo simulations and benchmark it against LYSO system.