Application of sinogram-affirmed iterative reconstruction technique in reducing radiation dose of chest CT examination

Objective: To discuss the value of sinogram-affirmed iterative reconstruction (SAFIRE) technique in reducing radiation dose in chest CT examination. Methods: Sixty-four patients underwent chest scan in Somatom Definition Flash CT with flash spiral model. Thirty-two patients underwent chest scan with 100 kV and SAFIRE (value=3) reconstruction (low-dose group), while other 32 patients underwent chest scan with 120 kV and filter back projection (FBP) reconstruction (control group). CT values and standard deviation (SD) of descending aorta and muscle at the level of epibronchial were measured. CT dose index (CTDI) and dose length product (DLP) were compared. Two radiologists assessed the image quality with 5-point scale. The degree of interobserver concordance was evaluated using Kappa test. Results: All images of both groups could satisfy diagnosis requirement, and the interobserver agreement for diagnosis acceptability was good (Kappa value was 0.795). CTDI was (3.23 ± 0.53)mGy in low-dose group, while (5.27 ± 1.07)mGy in control group. DLP was (114.96 ± 18.90)mGy · cm in low-dose group, while (184.40 ± 36.85)mGy · cm in control group. The radiation dose was (1.60 ± 0.26)mSv in low-dose group, while (2.58 ± 0.51)mSv in control group (all P<0.001). CT value of descending aorta was (39.90 ± 8.31)HU in low-dose group and (43.12 ± 7.43)HU in control group, SD was 13.61 ± 2.00 in low-dose group and 13.00 ± 3.39 in control group (all P>0.05). CT value of muscle at the level of epibronchial was (45.50 ± 11.85)HU in low-dose group and (48.23 ± 10.99)HU in control group, SD was 10.34 ± 1.55 in low-dose group and 8.53 ± 1.99 in control group (all P>0.05). Conclusion: SAFIRE can reduce radiation dose in adult chest CT scan on the condition that the image quality was not affected.