NOISE-PROPAGATION IN SPECT IMAGES RECONSTRUCTED USING AN ITERATIVE MAXIMUM-LIKELIHOOD ALGORITHM

The effects of photon noise in the emission projection data and uncert ainty in the attenuation map on the image noise in attenuation-correct ed SPECT images reconstructed using a maximum-likelihood expectation-m aximization algorithm were investigated. Emission projection data of a physical Hoffman brain phantom and a thorax-like phantom were acquire d from a prototype emission-transmission computed tomography (ETCT) sc anner being developed at UCSF. Computer-simulated emission projection data from a head-like phantom and a thorax-like phantom were also obta ined using a fan-beam geometry consistent with the ETCT system. The si mulation assumed a Tc-99m source, included collimator blurring but ign ored photon scatter. For each phantom, a region of interest (ROI) at t he centre of the reconstructed image was chosen for the purpose of noi se analysis. In all cases, the mean value (m) in the ROI approached a constant value after approximately 20 iterations. The standard deviati on (sigma) generally increased with the number of iterations. The rati o (sigma/m) was found to be inversely proportional to the square root of the total detected counts and proportional to the relative uncertai nty in the attenuation maps. These two noise components contributed in dependently towards the noise in the reconstructed image. In the ETCT system employing an x-ray tube for attenuation map acquisition, the un certainty in the reconstructed radionuclide distribution is limited ma inly by photon noise in the emission projection data. Our results are expected to be generally applicable to other emission-transmission sys tems, including those using external radionuclide sources for the acqu isition of attenuation maps.