Generation of realistic simultaneous cardiac and respiratory gated SPECT datasets using the 4D XCAT phantom and Monte Carlo simulation

Objectives The goal is to generate and evaluate a set of realistic simultaneous cardiac and respiratory gated SPECT datasets for use in the study of the effect of respiratory motion and the development of data acquisition and corrective image reconstruction methods for improved gated myocardial perfusion (MP) SPECT.

Methods To simulate both cardiac and respiratory motions using the new 3D XCAT (eXtended CArdiac Torso) phantom, one cycle of respiratory motion was divided into 24 while the cardiac beating motion over a cardiac cycle was divided into 48 time frames for each of the 24 respiratory phases. From this master 4D phantom dataset, almost noise-free projection datasets were generated separately from the heart, blood pool, lungs, liver, kidneys, stomach, gall bladder and the remaining body phantoms using the SimSET Monte Carlo code and angular response functions. To demonstrate their application, the individual organ projection datasets were scaled and combined to simulate a typical clinical 99mTc Sestamibi MP SPECT projection dataset with different gating schemes including no gating, 6 respiratory-gates with 8 cardiac-gates, and some combinations of them. Each projection set was reconstructed using the 3D OS-EM with and without attenuation correction (AC) using different attenuation maps. The reconstructed images were evaluated in terms of image artifacts and deviation from the corresponding phantom slices.

Results We demonstrated the flexibility and potential utility of the master dataset through a gated MP SPECT study with various dual gating schemes. The respiratory motion generated a signature artifactual decrease in both the anterior and inferior regions of the MP polar maps. While gating the SPECT data reduces the artifact significantly, AC with the gated attenuation map further reduced it.

Conclusions We conclude that the realistic 4D XCAT dual gating dataset provides a powerful tool in the study of the effects of cardiac and respiratory motions, gating schemes and correction methods for ECT/CT imaging.

Research Support NIH R01 EB000168, R01 HL06807