An experimental study of myocardial tomography artifacts caused by motion

Objective: To determine the relationship between the phantom motion and the resultant image artifact. Methods: A myocardial phantom was filled with 99TcmO-4solution. During imaging,the phantom was moved +- 5 mm, +- 10 mm, +- 20mm along the x, y and z axis, respectively at the angle of 45 degree, 90 degree, 135 degree sequentially. Each image was compared with its counterpart image acquired from no motion imaging and assessed by three experienced observers without knowledge of presence or absence of phantom motion. The results were analyzed by chi-square test. Results: When the phantom was moved 5 mm, there was no significant motion artifact. When the phantom was moved 10 mm or 20 mm, artifacts were obvious. The more moved, the more remarkable the artifacts. The artifacts showed unevenness of the radioactivity of the myocardial wall. In the short axis slices, the triangular patterns composed of the 'lumpy' and 'defect' areas in the images rotated to certain extent in position. In the horizontal and vertical long axis slices, the thickness of the lateral and septal wall, the anterior and inferior wall were decreased symmetrically. Conclusions: When the phantom was moved 10 mm or 20 mm, there were obvious motion artifacts. The artifacts have their specifics and can be recognized