Abstract
2096
Objectives Evaluate the performance of simultaneous 99mTc-MDP/123I-MIBG tumor imaging using fast Monte Carlo (MC) dual-radionuclide iterative reconstruction.
Methods Noise-free 99mTc and 123I SPECT projections were acquired separately using an anthropomorphic torso phantom modified to include a fillable tube around the lungs to mimic ribs. 99mTc and 123I projections were also acquired of a 1-cm spherical “lesion” at various distances from the collimator. Lesion-present data were generated by adding sphere projections to those of the torso phantom with two different lesion-to-background contrast ratios: marginally detectable (1.51/0.53 for 99mTc/123I), and conspicuous (3.20/1.63 for 99mTc/123I). The data were scaled to the total counts of a typical clinical study. 16 noise realizations were generated by adding Poisson noise to the projection data for each radionuclide. Dual-radionuclide projection data were generated by summing 99mTc and 123I sinograms. Image reconstruction was performed using: (1) SI-OSEM, OSEM without scatter correction (SC) using single-isotope (unmixed) data; (2) SI-MC-OSEM, OSEM with a fast MC-based SC using single-isotope data; (3) DI-OSEM, OSEM without SC using dual-isotope (mixed) data; (4) DI-MC-JOSEM, joint OSEM with a fast MC-based SC using dual-isotope data. Point spread functions were modeled in all the methods.
Results For each radionuclide, mean and standard deviation of lesion contrast were computed. Using SI MC-OSEM at 30th iteration as reference, DI-MC-JOSEM yielded 92.3%-99.0% contrast recovery for the lesions used in this study. This was significantly greater (p<0.000001) than the contrast associated with DI-OSEM.
Conclusions Simultaneous 99mTc-MDP/123I-MIBG oncologic imaging using DI-MC-JOSEM yielded comparable image quality compared to sequential 99mTc and 123I imaging and offers a promising alternative to current multi-visit multi-scan clinical standard.
Research Support NIH R21CA140771