The optimal protocols of low-dose CT with iterative reconstruction CT in PET/CT scan

2675 Objectives Iterative reconstruction (IR) has been successfully used in clinical CT scans and indicated to reduce image noise and improve image quality in low dose CT scan compared to filter back-projection (FBP). The purpose of this study is to evaluate the influence of low dose CT protocols with IR for attenuation correction in PET/CT scans. Methods CT images were acquired by the following parameters: 120 kV, six different tube current (120, 80, 60, 40, 20, 10 mA), reconstructed by FBP (CTFBP) and different weighted adaptive statistical iterative reconstruction (ASiR)(CTASiR). The cylinder phantom including four tubes filled with air, bone equivalent (40% K2HPO4), 4:1 and 8:1 18F-FDG solution was inserted in the middle of anthropomorphic cardio phantom (QRM phantom) to obtain PET images. And CTFBP and CTASiR were performed for attenuation correction on PET images. CT number, signal to noise ratio (SNR), standard uptake value (SUV) and modulation transfer function (MTF) were evaluated. Results The SUV in each region of interest (ROI) was not significantly different on PET images reconstructed by CTFBP(120 mA), CT FBP(10 mA) and different weighted CTASiR(10mA) (p>0.05). Noise on CTASiR images was significantly lower than CT FBP images at corresponding scanning parameters. Additionally, the noise level of CTASiR70(40mA), CTASiR90(40mA) and CTASiR100(40mA) was the same as standard images (CTFBP(120mA)). MTF10 % of CTASiR70(40mA), CTASiR90(40mA) and CTASiR100(40mA) was not significantly different compared to CTFBP(120mA) (p>0.05). Conclusions In this study, CT FBP(10 mA) and CT ASiR(10 mA) could be used for attenuation correction without a significant decrease in image quality, and dose reduction compared to CT120 mA was 91.6 %. CTASiR70(40mA), CTASiR90(40mA) and CTASiR100(40mA) could be used not only for attenuation correction but also anatomic localization, and dose reduction compared to CT120 mA was 66.7 %.