A new dynamic myocardial phantom for the assessment of left ventricular function by gated single-photon emission tomography

Abstract. Gated myocardial perfusion single-photon emission tomography (SPET) has been used for the measurement of left ventricular (LV) function and validated by means of comparison with other imaging modalities. We have designed a new dynamic myocardial phantom in order to validate the LV function as assessed by the use of gated myocardial perfusion SPET. The phantom consists of two half-ellipsoids (an endocardial surface and an epicardial surface) and a thorax. The myocardial space is filled with a radioactive solution. The endocardial surface moves continuously towards and away from the epicardial surface in the longitudinal axis to vary the LV volume [143 ml at end-diastole (ED), 107 ml at end-systole (ES)] and thickness (apex 8 mm at ED and 26 mm at ES, midplane 8 mm). The mean values of wall motion (WM) for the apical midplane region and the basal midplane region were 5 mm and 2 mm, respectively. Gated myocardial SPET was performed during 8 and 16 intervals. These projection data sets were processed using a Butterworth filter with an order of 5 and a critical frequency of 0.34 cycles/cm. LV function was calculated using the quantitative gated SPET (QGS) algorithm. The LV function values estimated by gated SPET during 16 intervals [22% for ejection fraction (EF), 3.7 mm for WM of the apical midplane, 1.7 mm for WM of the basal midplane] closely resembled actual LV functions [25% for EF, 5 mm for WM of the apical midplane, 2 mm for WM of the basal midplane]. However, the estimated values during 8 intervals were smaller than those during 16 intervals (19% for EF, 3.3 mm for WM of the apical-midplane, 1.1 mm for WM of the basal-midplane). The estimated LV volumes closely correlated with the actual volumes (r=0.99 for 16 intervals, r=0.95 for 8 intervals). Utilizing this phantom, LV function estimated using gated myocardial SPET can be compared with actual values.