A simulation study of the effect of gating scheme on respiratory motion bluffing in FDG Lung PET

In this study, different respiratory gating schemes in FDG Lung PET were studied using a modified 4D NURBS based cardiac torso (NCAT) phantom and Monte Carlo simulation methods. A sphere-shaped lesion with 7 mm in diameter was placed at the bottom of the right lung, where the magnitude of respiratory motion was the largest. Projection data of the 4D NCAT phantom with the lung lesion and respiratory motion and using a PET system were simulated using the SimSET Monte Carlo code. The effects of the respiratory motion under several gating schemes were studied. A total of 240 projection time frames were generated over one respiratory cycle (∼5 seconds) and were divided and summed into 3, 4, 6, 8 gated frames. The sets of gated frames were moved along the respiratory cycle to form different gating schemes. The FWHMs of the reconstructed lesion image were determined along the direction of movement due to respiratory motion under different gating schemes. The results demonstrate how the number of gates and their placement in a respiratory cycle would affect the blurring of a lung lesion along the direction of movement. Specifically, a small number of gates give more lesion blurring within each gate and the placement of the gates within the respiratory cycle is more important to minimize the average blurring among all gates. How ever, a large number of gates result in higher level of image noise within each gate. Further study is needed to determine the number and placement of respiratory gates in lung PET studies.