The effect of energy on scintillation camera intrinsic spatial resolution

Accurate activity quantification in Nuclear Medicine Imaging is a prerequisite for accurate absorbed dose calculations and quantitative biodistribution studies. Monte Carlo modelling is a useful tool for studying quantitative accuracy in scintillation camera imaging. In order to simulate spatial response, a value for the intrinsic spatial resolution is needed. In this study we have by means of experimental measurements and simulations investigated the intrinsic spatial resolution for the photopeaks of ^99mTc, ¹¹¹In and ¹³¹I. This was to enable the validation of the Monte Carlo code GATE. Experimental images were acquired on a 0.95 cm thick NaI(Tl) crystal Philips FORTE and a 1.59 cm thick NaI(Tl) crystal Philips SKYLight scintillation camera. A 1 mm slit phantom was placed in close proximity to the crystal and a source holder, constructed so that only a narrow beam of radiation was incident on the slit, was placed 1.5 m from the crystal. GATE was used to simulate projection data under the same conditions as the experiments. In the case of ¹¹¹In the two photopeaks (171.3 keV and 245.4 keV) were simulated separately. By comparing the known intrinsic spatial resolution of the crystal caused by multiple scattering of photons in the crystal to the total intrinsic spatial blurring, the intrinsic spatial component caused by electronic processing was estimated. Our data show that the intrinsic spatial blurring introduced by the electronics is dependent on the energy of the photons being imaged and can be accurately estimated. The difference between the intrinsic spatial resolution value for ^99mTc (140.5 keV) and ¹³¹I (364.5 keV) is 27% for the thin crystal and 21% for the thick crystal.