Effect of adaptive statistical iterative reconstruction algorithm on the imaging quality in low-dose spectral CT scanning of the liver

Objective To investigate the value of the adaptive statistical iterative reconstruction (ASIR) algorithm for reducing the radiation dose and optimizing the image quality in the low-dose spectral CT scanning in GSI (Gemstone spectral imaging) of the liver. Methods A total of 60 patients who underwent hepatic spectral CT scanning in GSI were enrolled in this study. The patients were randomly divided into two groups according to priority with 30 cases per group. Low-dose spectral CT scanning was used for group A, and images were reconstructed by ASIR 0 and 50%, marked as A1 and A2. Group B was scanned with conventional dose of spectral CT, and images were reconstructed by Filtered back projection (FBP). Effective doses (E) for each group were calculated. Image quality was assessed by two radiologists, and the radiation doses were compared between groups A and B. Results All image quality of each group were good enough for clinical diagnosis. E for group A and B were (3.2±0.2) and (5.8±0.2) mSv, respectively. There was statistical difference with image noise between group A and B(Z=-6.784, P 0.05). There were no statistical differences of the image quality score between groups A1, A2 and B (F=102.38, 105.768, P<0.05). Conclusions ASIR combined with low-dose spectral CT scanning was helpful to reduce radiation dose and could obtain better image quality in hepatic CT examination. Key words: Liver; Spectral imaging; Adaptive statistical iterative reconstruction (ASIR); Radiation dose; Tomography, X-ray computed