REGION OF INTEREST EVALUATION OF SPECT IMAGE-RECONSTRUCTION METHODS USING A REALISTIC BRAIN PHANTOM

A realistic numerical brain phantom, developed by Zubal et al, was use d for a region-of-interest evaluation of the accuracy and noise varian ce of the following SPECT reconstruction methods: 1) Maximum-Likelihoo d reconstruction using the Expectation-Maximization (ML-EM) algorithm; 2) an EM algorithm using ordered-subsets (OS-EM); 3) a re-scaled bloc k iterative EM algorithm (RBI-EM); and 4) a filtered backprojection al gorithm that uses a combination of the Bellini method for attenuation compensation and an iterative spatial blurring correction method using the frequency-distance principle (FDP). The Zubal phantom was made fr om segmented MRI slices of the brain, so that neuro-anatomical structu res are well defined and indexed. Small regions-of-interest (ROIs) fro m the white matter, grey matter in the center of the brain and grey ma tter from the peripheral area of the brain were selected for the evalu ation. Photon attenuation and distance-dependent collimator blurring w ere modeled. Multiple independent noise realizations were generated fo r two different count levels. The simulation study showed that the ROI bias measured for the EM-based algorithms decreased as the iteration number increased, and that the OS-EM and RBI-EM algorithms (16 and 64 subsets were used) achieved the equivalent accuracy of the ML-EM algor ithm at about the same noise variance, with much fewer number of itera tions. The Bellini-FDP restoration algorithm converged fast and requir ed less computation per iteration. The ML-EM algorithm had a slightly better ROI bias vs. variance trade-off than the other algorithms.