Abstract
Background: Digital anthropomorphic phantoms have gradually gained an important role in nuclear medicine imaging. The aim of this study was to generate digital phantom models simulated with cardiac and respiratory motions and to evaluate motion-induced artifact using a quantitative software program developed with artificial intelligence (AI) technology in myocardial perfusion single-photon emission computed tomography (SPECT) images.
Methods: A digital anthropomorphic torso phantom and Monte Carlo simulation program were used to generate the projection image. We prepared electrocardiographic- and respiratory-gated models with cardiac and respiratory motions in the diaphragmatic motion of 0-50 mm and chest anterior-posterior (AP) expansion of 0-30 mm. The phantoms had 99mTc radiopharmaceuticals embedded based on a normal male subject. The projection data were reconstructed with filtered back projection method. The 17-segment polar map without attenuation correction was created. Myocardial count ratio was calculated in the anterior and inferior walls to the anterolateral wall on the polar map.
Results: The low myocardial perfusion distribution was clearly observed in the basal to mid anterior and inferior walls, when the diaphragmatic motion and AP expansion were increased. The anterior and inferior count ratios showed significant decrease with diaphragmatic motion (≥30 mm) compared to without diaphragmatic motion (0.71±0.05 vs. 0.76±0.06; p<0.0001, 0.67±0.04 vs. 0.69±0.04; p=0.038, respectively). The anterior and inferior motion artifacts could not be detected by the quantitative software program in the diaphragm motion of ≤20 mm and AP expansion of ≤12 mm.
Conclusions: The cardiac and respiratory motion-induced artifacts were characterized as low anterior and inferior myocardial counts. Therefore, motion artifacts should be considered when we observe this sort of count decreases on the reconstructed images and polar maps. However, we could exclude the mild motion-induced artifacts by using the quantitative software program integrated with AI technology.
