Compensation for Patient and Detector Scatter and Crosstalk Contamination in 111In SPECT Using Fast Monte Carlo-based Iterative Reconstruction

We have developed a fast Monte Carlo (MC)-based joint ordered-subset expectation maximization (JOSEM) iterative reconstruction algorithm, MC-JOSEM. A phantom was scanned with CT and SPECT to compare quantitative imaging performance of MC-JOSEM with that of a triple-energy-window method (TEW) in which estimated scatter was also included additively within JOSEM, TEW-JOSEM. We acquired very-high-count projections of three 5.5-cc spheres of ¹¹¹In taped in different locations in a water-filled torso/spine/lung phantom; 14 scans were then obtained with ¹¹¹In only in the background for generating synthetic images of spheres in background. MC scatter estimates, which were used by MC-JOSEM, were computed once after 5 iterations (60 projections, 10 subsets) of TEW-JOSEM. Images after 36 iterations of both TEW-JOSEM and MC-JOSEM were obtained and counts were measured within volumes-of-interest centered on the spheres in all sphere+background and background-only image sets. The bias ranged from 7.7% to 14.1% for TEW-JOSEM, and from 8.9% to 11.3% for MC-JOSEM, while the precision was significantly better for MC-JOSEM (5.4% to 11.2%) than for TEW-JOSEM (9.2% to 17.9%). MC-JOSEM is a promising approach for quantitative SPECT.