Feasibility study on half-dose OSEM reconstruction in gated myocardial perfusion imaging with single photon emission tomography

Abstract

Objective To evaluate the feasibility and values of half-dose and ordered subset expectation maximization(OSEM) reconstruction in single photon gated myocardial perfusion imaging(G-MPI). Methods Patients with suspected or confirmed coronary artery diseases(CAD) underwent G-MPIs were divided into two groups.One group underwent SPECT G-MPI with routine 25 s acquisition time, and 50 s acquisition time for another group. After completion of acquisitions, raw-data in each group were respectively reconstructed with filtered back projection(FBP) and OSEM. Each subgroup images were evaluated with five-point scores for image-quality evaluation, and five-point summed scores were calculated on base of left ventricle 17 segment model. Image quality scores, summed stress score(SSS) or summed rest score(SRS) and left ventricle ejection fraction(LVEF) were respectively calculated and compared for FBP or OSEM images of each subgroup. Percentage of excellent images(score 4) was calculated and compared for OSEM images of routine acquisition time group or doubling time group. Results For half-dose and routine 25 s acquisition time group, the FBP image quality scores were less than those for OSEM group(t=-3.72, P 0.05). For half-dose and doubling acquisition time group, there were no significant differences in image quality scores, summed radioactive scores or LVEF between FBP and OSEM images(P all>0.05). There was no significant differences in percentage of excellent images(score 4) between routine 25 s acquisition time group and doubling acquisition time group(χ2=1.00, P>0.05). Conclusions Under the circumstance of half-dose injection, the image qualities reconstructed with OSEM are superior to those reconstructed with FBP for routine 25 s acquisition time group, and images reconstructed with FBP may overestimate the severity. FBP or OSEM image qualities are comparable for doubling acquisition time group, and there are no significant differences in summed radioactive score and LVEF. So half-dose, routine acquisition time and OSEM reconstruction provide good image qualities, and no need for additional acquisition time, and radiation doses for patients are also simultaneously reduced because of low injection dose, which has good feasibility and values in the clinical applications. Key words: Half-dose; Ordered subset expectation maximization; Single photon emission tomography; Myocardial perfusion imaging