The Analysis of Correction Effect of Motion Artifacts in PET Images Using Respiratory Gating System

Abstract

PET is used effectively for biochemical or pathological phenomena, disease diagnosis, prognosis determination after treatment, and treatment planning because it can quantify physiological indicators in the human body by imaging the distribution of various biochemical substances. However, since respiratory motion artifacts may occur due to the movement of the diaphragm due to breathing, we would like to evaluate the practical effect by using the a device-less data-driven gated technique called Motion Free (DDG) with the phase-based gating correction method called Q.static scan mode. In this study, images of changes in moving distance (0 cm, 1 cm, 2 cm, 3 cm) and degree of slope (0°, 2.5°, 5°, 10°) were acquired using a breathing-simulated moving phantom. The diameters of the six spheres in the phantom are 10 mm, 13 mm, 17 mm, 22 mm, 28 mm, and 37 mm, respectively. According to SUVmax and SUVmean measurements, when DDG was applied based on the moving distance, the average SUVmax of the correction effect by the moving distance was improved by 1.92, 2.48, 3.23 and 3.00, respectively, and the average SUVmean was improved by 0.04, 0.72, 1.36 and 1.49, respectively. When DDG was applied based on the diameter of the phantom spheres, the average SUVmax of the correction effect by the moving distance was improved by 2.37, 2.02, 1.44, 1.20, 0.42 and 0.52, and the average SUVmean was improved by 1.36, 1.23, 0.99, 0.82, 0.57 and 0.43, respectively. The correction effect by the degree of slope also improved the average SUVmax when DDG was applied to 2.93, 2.87, 1.61 and 1.45 respectively, and the average SUVmean to 1.61, 1.27, 0.8 and 0.95 respectively, based on the degree of slope. when DDG was applied based on the diameter of the phantom sphere, the average SUVmax was improved by 0.77, 1.73, 1.84, 2.56, 3.46 and 2.93 respectively and the average SUVmean was improved by 0.56, 0.44, 1.19, 0.98, 1.66 and 2.13, respectively. In the visual comparison due to the change of the moving distance and the degree of slope, it was found that the more the moving distance increases, the greater the degree of slope increases, the greater the correction effect of DDG. In the case of the area ratio, Based on the image with a moving distance of 0 cm, when moving distance was 1 cm, 2 cm and 3 cm, the difference between applying and not applying DDG was 5%, 12%, and 18%, respectively. Based on an image with degree of slope of 0°, when degree of slope was 2.5°, 5° and 10°, the difference between applying and not applying DDG was 6%, 7%, and 7%, respectively. In the case of the average density ratio, Based on an image with a moving distance of 0 cm, when moving distance was 1 cm, 2 cm and 3 cm, the difference between applying and not applying DDG was 3%, 6%, and 8%, respectively. Based on an image with degree of slope of 0°, when degree of slope was 2.5°, 5° and 10°, the difference between applying and not applying DDG was 3%, 4%, and 5%, respectively. In the comparison of FWM, FWHM, and FWTM, Based on the image with a moving distance of 0 cm, when moving distance was 1 cm, 2 cm and 3 cm, the difference between applying and not applying DDG was 1.6%, 9% and 9.4% in FWM, 4.4%, 7.6% and 6.4% in FWHM, 12.1%, 23.4% and 30.5% in FWTM respectively. Based on an image with degree of slope of 0°, when degree of slope was 2.5°, 5° and 10°, the difference between applying and not applying DDG was 3.0%, 3.2% and 11.7% in FWM, 2.7%, 5.0% and 2.3% in FWHM, 12.2%, 6.3%, and 7.5% in FWTM respectively. It means there is a quantitative correction effect when applying DDG. This shows that the overall experiment has a correction effect when applying DDG.