A study of the effect of cardiac gating in myocardial SPECT using the 4D NCAT

Abstract

The purpose of the study is to investigate the effect of cardiac gating in myocardial SPECT obtained from different gating schemes and image reconstruction methods using the 4D NURBS-based cardiac-torso (NCAT) phantom. We generated 144 phantoms over a cardiac cycle modeling a typical Tc-99m Sestamibi study. The cardiac cycle was divided into different gating schemes (8, 12, 16, and 24 gates) by summing combinations of these phantoms. For each gate, we generated projection data using an analytical projector that included the effects of attenuation, collimator-detector response and scatter, and reconstructed using different image reconstruction methods including FBP, OS-EM (with and without attenuation, scatter, and collimator-detector response corrections), and a 4D MAP-RBI-EM algorithm. The amount of motion blur (measured by mean square error) and percentage fluctuation of the reconstructed images were evaluated as measures for temporal resolution and noise level. The ejection fraction was also calculated to further evaluate the different gating schemes and reconstruction methods. The results show a dramatic improvement in temporal resolution in gated versus ungated images. The level of noise was found to increase significantly with an increase in the number of gates. Using a smaller number of gates was found to have a higher variation in estimating the ejection fraction. For each gating scheme, FBP and OS-EM without correction were found to have the poorest temporal resolution and the highest level of image noise while OS-EM with corrections was found to significantly improve temporal resolution and reduce image noise. We conclude that the 4D NCAT phantom is a useful tool in the study of optimal gating schemes and 4D image reconstruction methods for improved gated myocardial SPECT.