Optimization of reconstruction parameters using a multi-focus fan beam collimator in myocardial perfusion single photon emission computed tomography study.

Abstract

The aim of this study was to optimize the reconstruction parameters for ordered subset conjugate gradient minimization (OSCGM) reconstruction using a multifocus fan beam collimator in myocardial perfusion single photon emission computed tomography (SPECT). We attempted to validate the following performance of OSCGM reconstruction parameters (iteration and subset). SPECT images were acquired using a dual-head gamma camera with IQ mode acquisition systems from a RH-2 cardiac phantom containing a 99m-Tc solution. The performance was evaluated using reconstruction parameters (product of subset and iteration: SI) with image contrast, LV volume [using quantitative perfusion SPECT (QPS)], root mean square uncertainty (RMSU), and normalized mean squared error (NMSE). The best results (contrast, uniformity, LV volume, and NMSE) were found for SI: 30. LV volume indicated the true volume for subset: 1 and iteration: 30, and LV volume was underestimated by 10% for iteration >20, and subset >1. The results of this myocardial perfusion SPECT study suggest the optimal OSCGM reconstruction parameter to be subset: 1 and iteration: 30 using a multifocus fan beam collimator.