TY - JOUR T1 - Comparison of Left Ventricular Ejection Fraction Calculation Methods between CT and SPECT with a Dynamic Cardiac Phantom JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 24 LP - 24 VL - 58 IS - supplement 1 AU - Tinsu Pan AU - Guang-Uei Hung Y1 - 2017/05/01 UR - http://jnm.snmjournals.org/content/58/supplement_1/24.abstract N2 - 24Objectives: Most comparisons in ejection fraction (EF) calculation methods between two different imaging modalities were performed on the same patient at two different times without knowing the true EF, end-systolic volume (ESV) or end-diastolic volume (EDV). We utilized a dynamic cardiac phantom with programmable ESV, EDV and heart rate to remedy this shortcoming, and compared cardiac CT, which has high spatial resolution and lower temporal resolution of 175 msec for coronary artery imaging of the patient whose heart rate ≤65 bpm, and cardiac SPECT, which has high temporal resolution but poor spatial resolution, in the aspects of EF calculation, as well as ESV and EDV estimates.Methods: A Data Spectrum Dynamic Cardiac Phantom was programmed for EF=30 and 60 for ESV=91 and 52 cc (same EDV=130 cc) respectively for the heart rates of 40 to 100 bpm in increment of 20 bpm. The myocardium walls of the phantom were with Tc-99m solution for the SPECT (Siemens Symbia), and with iodine contrast solution for the 64-slice CT (GE) acquisitions, respectively. The parameters for SPECT were low energy high resolution collimator, 64 angles over 180 degree, 128x128 matrix, zoom=1.23, and 16 gates. The parameters for CT were 0.35 sec gantry rotation, 10 phases, 512x512 matrix, and 2.5 mm slice thickness. Each scan was repeated twice in SPECT and three times in CT. Cedars quantitative gated SPECT and MIM software were used for the SPECT and CT EF calculations, as well as ESV and EDV estimates, respectively.Results: In SPECT, there were -1.3±1.9% and -1±1.1% in error for the EF=30 and EF=60, respectively. Both ESV and EDV estimates were -19% and -17% in error, respectively for EF=30, and -20% and -18.5% in error, respectively for EF=60. There was no heart rate dependence for the results in SPECT. In CT, for EF=30, there were -1.1±0.4%, -1.7±0.5%, -2.8±0.2% and -4.4±0.7% in increasing error for the 45, 60, 80, and 100 bpm, respectively; and for EF=60, there were -1.8±0.6%, -2.2±0.8%, -4.6±0.3%, and -7.4±1.8% in increasing error for the 45, 60, 80, and 100 bpm, respectively. Both ESV and EDV estimates from CT were 1.4% and -2.8% in error, respectively for EF=30, and 2.6% and -8.3% in error, respectively for EF=60.Conclusion: SPECT was accurate in EF calculation but could be off by 20% in both ESV and EDV estimates likely due to its poor spatial resolution. CT was accurate in EF calculation for the heart rates ≤ 65 bpm, and was under-estimate in EF calculation for the higher EF=60 or higher heart rates of 80 and 100 bpm from both over-estimate of ESV and under-estimate of EDV, likely due to its lower temporal resolution not capable of freezing the cardiac motion. However, CT was more accurate in ESV and EDV estimates than SPECT. ER -