Impact of Preset and Postset Adaptive Statistical Iterative Reconstruction-V on Image Quality in Nonenhanced Abdominal-Pelvic CT on Wide-Detector Revolution CT.
Zheng Zhu,Yanfeng Zhao,Xinming Zhao,Xiaoyi Wang,Weijun Yu,Mancang Hu,Xuan Zhang,Chunwu Zhou
DOI: https://doi.org/10.21037/qims-19-945
2021-01-01
Quantitative Imaging in Medicine and Surgery
Abstract:BACKGROUND:Adaptive statistical iterative reconstruction-V technique (ASIR-V) is usually set at different strengths according to the different clinical requirements and scenarios encountered when setting scanning protocols, such as setting a more aggressive tube current reduction (defined as preset ASIR-V). Reconstruction with ASIR-V is useful after scanning using image algorithms to improve image quality (defined as postset ASIR-V). The aim of this study was to investigate the quality of images reconstructed with preset and postset ASIR-V, using the same noncontrast abdominal-pelvic computed tomography (CT) protocols in the same individual on a wide detector CT.METHODS:We prospectively enrolled 141 patients. The scan protocols in Groups A-E were 0%, 20%, 40%, 60%, and 80% preset ASIR-V, respectively, in the 256 wide-detector row Revolution CT (GE Healthcare, Waukesha, WI, USA). Each group was further divided into 5 subgroups with 0%, 20%, 40%, 60%, and 80% postset ASIR-V, respectively. The 64-detector Discovery 750 HDCT (GE, USA) was used for Group F as a control group, using 0%, 20%, 40%, 60%, and 80% ASIR, respectively. Image noise was measured in the spleen, aorta, and muscle. The CT attenuation and image noise were analyzed using the paired t-test; analysis of variance and post hoc multiple comparisons were made using the Student-Newman-Keuls (SNK) method.RESULTS:The CT attenuation in Groups A-F exhibited no significant difference between subgroups in three organs (P>0.05). Only with increasing preset ASIR-V% (Groups A to E), did the image noise decrease, except in Group B in the aorta and muscle (NoiseB > NoiseA, PmuscleA&B=0.233, PaortaA&B=0.796). Only with increasing postset ASIR-V or ASIR% (Groups A and F), did the image noise decrease in the three organs. After preset and postset ASIR-V were combined, with preset ASIR-V% being equal to postset ASIR-V%, the image become similar to the corresponding preset ASIR-V part with the line of postset ASIR-V 0% (baseline of each group). When preset ASIR-V% was greater than the postset ASIR-V%, the image noise was higher than the baseline of each group. When preset ASIR-V% was less than the postset ASIR-V%, the image noise was lower than the baseline of each group. The radiation dose from B to E decreased from 11.2% to 57.1%. The CT dose index volume (CTDIvol) and dose length product (DLP) in Group F were significantly higher than those in Group A.CONCLUSIONS:Using both preset and postset ASIR-V allows dose reduction, with a potential to improve image quality only when postset ASIR-V% is higher than or equal to preset ASIR-V%. The image quality depends on postset ASIR-V%, whereas the decrease of radiation dose depends on preset ASIR-V%.