Document Type: Original Article
Department of Medical Physics, Tarbiat Modarres University, Tehran, Iran
Division of Nuclear Medicine, Geneva University Hospital, Geneva, Switzerland
Department of Biostatistics, Tarbiat Modarres University, Tehran, Iran
Introduction: Due to the large number of PET systems available in the market, it is not very easy to decide about the scanner of choice. Conducting a research in order to compare all different PET scanners is very time consuming and expensive and practically impossible. However, such comparison may be conducted using PET simulators. In this study, the performance of 6 different PET scanners in cardiology is evaluated using a dedicated PET simulator. In this study only the design of the system were evaluated. Methods: Activity and attenuation phantoms were produced using 4D-NCAT phantom. ECAT EXACT HR+, ECAT 953B, ECAT 966, ECAT ART, GE Advance and 16HI-REZ scanners were simulated using Eidolon PET simulator and the output sinograms were reconstructed using STIR software. The reconstructed images were processed using Interview software installed on the Mediso cardiac imaging system. Counts of pixels determined by ROI were used to drawn curves and then the correlations of these curves calculated using SPSS. Results: True coincidences 2D was 4651791±5900 for ECAT 966, 4651965±5660 for ECAT ART, 4742731±5328 for ECAT EXACT HR+, 6018435±5167 for ECAT 953B, 6566769±4734 for GE Advance and 6846339±51850 for 16HI-REZ. Resulted correlations calculated for these scanners were 0.806, 0.795, 0.718, 0.858, 0.726 and 0.896 respectively. Conclusion: There was a considerable different in scatter fractions of different scanners. Curves drawn using count of pixels determined by ROI and correlations of these curves showed differences in performances of scanners in cardiology. The results showed that the 16HI-REZ scanner is the best scanner of the six scanners for simulating of cardiac PET images.