The effect of β + energy on performance of a small animal PET camera

Abstract

The effective spatial resolution of a positron emission tomography (PET) scanner is determined in part by the initial energy of the positron, which is a function of the radionuclide. For F-18 ( E max=0.633 MeV) the mean positron range in water is small (0.6 mm). However, many other useful positron-emitting nuclides have higher energies, for example Ga-68 ( E max=1.899 MeV, mean range 2.9 mm) has one of the highest. The performance of a non-rotating, 16 module high density avalanche chamber (quad-HIDAC) small animal PET scanner was measured using both F-18 and Ga-68 to represent the extremes of high and low positron energy. The count rate performance—scatter fraction and noise-equivalent count rate (NEC)—were measured for both isotopes. Data were also collected for a spatial resolution phantom with rectangular arrays of holes of diameter 2.0, 1.5, 1.0 and 0.5 mm with the centres separated by 4.0, 3.0, 2.0 and 1.0 mm respectively. The NEC, measured for both 70 and 200 cm 3 cylindrical phantoms, was approximately linear up to 30 MBq, but shows a rapid drop-off above this value. The spatial resolution phantom showed that 1 mm objects are just resolved with F-18, but none of the targets are resolved for Ga-68. In conclusion, spatial resolution is dominated by the choice of isotope down to 1 mm, with sensitivity and count-rate data being largely independent of positron range.