Phantom study of the influence of iterative algorithm on image quality in ultra-low-dose CT scan of lung

Objective To explore the influence of iterative algorithm on image quality in ultra-low-dose CT scan of lung. Methods The thoracic model was scanned using different parameter combinations. The tube voltage was chose with 80, 100 kV in ultra-low dose group and the mAs was selected by 10, 15, 20, 25 and 30 mAs. The control group selected the parameters of 120 kV, 30 mAs. All the images were reconstructed with filtered back projection (FBP group) and iterative algorithm (IR group). The image noise and effective dose (E) were compared. Results When tube current and voltage were constant, the image noise of IR group was lower than that of FBP group, and the difference was statistically significant (t=1.102–8.070, P 0.05). Compared with the conventional low dose group, the E of the groups of (80 kV, 25 mAs), (80 kV, 30 mAs), (100 kV, 10 mAs), (100 kV, 15 mAs), (100 kV, 20 mAs), (100 kV, 25 mAs), (100 kV, 30 mAs) was decreased by 75.9%, 71.0%, 79.8%, 70.4%, 60.3%, 50.2%, 40.0%, respectively. Conclusions The image quality of the ultra-low dose protocol (100 kV, 10mAs) with iterative algorithm is similar to that of the conventional low dose with FBP group, and the radiation dose could be significantly reduced. Key words: Tomography, X-ray computed; Lung; Radiation dosage; Image quality; Iterative algorithm