Abstract: Objective: To compare the performance difference on image quality between sinogram-affirmed iterative reconstruction (SAFIRE) and filtered back projection （FBP）in low-dose lung CT screening. Methods: A total of 304 patients receiving annual low-dose lung CT screening were examined by a dual-source CT system at 120 kV with reference tube current of 40 mA·s. Six image serials were reconstructed, including one dataset of FBP and five datasets of SAFIRE with different reconstruction strengths from 1 to 5 (S1 to S5). Image noise and subjective scores of image noise were recorded. Images artifacts and overall image quality were also assessed by two radiologists. Results: The mean weight and the body mass index of patients were （66.3±12.8） kg and 23.4±3.2, respectively. The mean dose-length product was (95.2±30.6） mGy·cm, and the effective dose was （1.6±0.5） mSv. The observation agreement for image noise grade, artifact grade and the overall image quality was in good consistency (r=0.785, 0.595 and 0.512). Among the overall six datasets, both the scores of measured mean objective image noise and the subjective image noise of FBP were the highest, and the image noise decreased with the increasing of SAFIRE reconstruction strength. The datasets of S3 obtained were the highest quality scores for image. Conclusions: Compared with FBP, SAFIRE reconstruction may significantly improve image quality in low-dose lung CT screening and SAFIRE with reconstruction strength 3 is a pertinent choice for low-dose lung CT.

Key words: Lung, Computed tomography, Image reconstruction, Radiation dose