Abstract
The effect of sampling rate on the measurement of left-ventricular (LV) ejection fraction (EF), peak ejection rate (PER), and peak filling rate (PFR) was evaluated by computer simulation and clinical equilibrium blood pool imaging undertaken at rest and during excercise. In this simulation study, the effects of counting and fluctuation of R-R interval on statistical variation of ideal left-ventricular curve (assumed to consist of three harmonics) were taken into account by assuming Poisson and Gaussian distribution, respectively. The LV curves obtained with temporal resolution of 10 msec were added to make LV curves with temporal resolutions of 20, 30, and 40msec from which systolic and diastolic parameters were caluculated. On the other hand, LV curves with temporal resolution of 10 msec, undertaken at rest, and those which 20 msec, undertaken during excercise, were used to make LV curves with temporal resolution of 40msec. By observing changes in parameters measured with different temporal resolutions, we concluded that: a) sampling rate of 20-40 msec is recommended for rest study ; and b) sampling rate of 20msec is required for excercise study.
