Exercise first-pass radionuclide ventriculography provides valuable diagnostic and prognostic information in patients with coronary artery disease. In this procedure, motion correction of the images is commonly performed using a second external point source attached to the chest wall during exercise (dual-isotope method). Recently, a motion correction algorithm without an external point source (single-isotope method) was developed and the results compared with those of the dual-isotope method. METHODS To examine the accuracy of the motion correction method, a phantom study was performed using a moving cardiac phantom with a motion speed of up to 169 cycle/min and motion amplitude up to 6 cm. Count fluctuation in the phantom region by motion was calculated as a coefficient of variation (CV). In the clinical study, time-activity curves of the left ventricular phase were created for quantitative assessment of variation as CV values of the ejection fraction in the central five cardiac cycles after correction by the two methods during exercise radionuclide ventriculography in 17 patients. RESULTS In the moving phantom, both the single- and dual-isotope methods reduced the CV values less than 10%. In the clinical study, the single-isotope method provided less CV value of ejection fraction (9.8% +/- 5.6%) than the dual-isotope method (24.8% +/- 10.5%) (p < 0.01), indicating less individual variation of ejection fraction values. CONCLUSIONS These data indicate that object motion can be accurately corrected in the moving phantom by both single- and dual-isotope methods. In clinical studies, the single-isotope method is more accurate.