Impact of respiratory motion correction on diagnostic accuracy of myocardial perfusion SPECT

Abstract

This study aims to investigate the impact of respiratory motion correction on the diagnostic accuracy in myocardial perfusion SPECT. SPECT/CT data were acquired using an anthropomorphic phantom. The lungs were given an exponential sine-wave respiratory pattern resulting in heart motion of 1.5cm and 2.7cm in the cranio-caudal direction at the level of the diaphragm for the simulation of normal and deep breathing respectively. Defects were inserted into the myocardial wall. Motion correction was applied using an image based method with transformation parameters derived by a non-rigid registration algorithm. Images were evaluated by nuclear medicine physician. Respiratory motion correction eliminated artifactual defects present in the images, resulting in uniform uptake and reduction of motion blurring especially in the inferior and anterior walls. Increased defect contrast and thinner appearance of the heart wall were observed. The extent of improvement depended on the amplitude of the motion displacement; Although the impact of motion correction on the detection task was not as apparent for 1.5cm as with 2.7cm motion, in both cases correction recovered the correct defect size and contrast while rectifying the artifactual decrease of uptake. Finally, a combination of motion correction with attenuation correction enhanced these findings. Respiratory motion correction reduces motion artefacts in SPECT images resulting in improved defect detectability in myocardial perfusion imaging.