CT理论与应用研究2025，Vol.34Issue(3)：359-368,10.DOI:10.15953/j.ctta.2025.001
低管电压联合深度学习图像重建算法在降低胸腹部联合增强CT辐射剂量的价值
Value of Low Tube Voltage Combined with Deep Learning Image Reconstruction Algorithm to Reduce Radiation Dose in Combined Thoracoabdominal Enhanced CT
摘要
Abstract
Objective:To investigate the effect of low tube voltage combined with deep learning image reconstruction(DLIR)on radiation dose reduction and maintaining image quality in combined chest and abdominal enhanced CT scans.Methods:(1)Phantom study.To determine the feasibility of combining low tube voltage with deep learning algorithms for low-contrast resolution,Catphan 500 phantoms were scanned under two different conditions.The optimization group used a low tube voltage(80kV)combined with DLIR for scanning and image reconstruction,while the routine group used a 120kV tube voltage combined with adaptive statistical iterative reconstruction V(ASiR-V).This study aimed to determine the effectiveness of the optimization group using a low dose(noise index,NI>9)compared with the routine group using a routine dose(NI=9).(2)Prospective study.A total of 160 patients who underwent routine chest and abdominal enhanced CT scans were prospectively collected and randomly divided into a low-dose optimization group and routine-dose group,with 149 patients ultimately enrolled(61 in the low-dose optimization group and 88 in the routine-dose group).Based on the results of the phantom study,the low-dose optimization group used the optimized condition with NI set to the optimal value,whereas the routine-dose group used the routine condition with NI=9.Radiation doses were recorded and calculated for both groups,and image quality was subjectively and objectively evaluated.Results:The low-dose optimization group using NI=12 achieved an equivalent low-contrast resolution capability to the routine-dose group with NI=9.The effective dose in the low-dose optimization group(9.56±2.34)mSv was significantly lower than that in the routine-dose group(17.82±5.22)mSv.The liver and aorta attenuation values in the low-dose optimization group were significantly higher than those in the routine-dose group,and the CNR and SNR values in the liver and aorta were also significantly higher.The spatial resolution of the aorta,common hepatic artery,and portal vein and the display of small vessels/bronchi were all superior in the low-dose optimization group compared with the routine-dose group.Conclusion:The combination of a low tube voltage and deep learning image reconstruction algorithm can ensure equivalent or even higher image quality while reducing radiation dose,providing a feasible solution for optimizing radiation dose in large-scale CT scans such as the combined thoracoabdominal enhanced CT.关键词
计算机体层摄影/深度学习图像重建算法/低管电压/辐射剂量/胸腹部联合CT扫描Key words
computed tomography(CT)/deep learning image reconstruction algorithm/low tube voltage/radiation dose/combined chest and abdomen CT scan分类
医药卫生引用本文复制引用
綦维维,程瑾,陈楚韩,安备,刘晓怡,付玲,王屹..低管电压联合深度学习图像重建算法在降低胸腹部联合增强CT辐射剂量的价值[J].CT理论与应用研究,2025,34(3):359-368,10.基金项目
北京大学人民医院发展基金专项(RD-2022-25). （RD-2022-25）
