基于Karl迭代重建的超低剂量CT对肺密度与肺容积定量分析的影响  

Effect of ultra-low dose CT based on Karl iterative reconstruction on quantitative analysis of lung density and lung volume

作　　者：朱彦瑾 黄晓旗 王雷 赵凡惠 郭佑民[2] 李建龙[1] ZHU Yanjin;HUANG Xiaoqi;WANG Lei;ZHAO Fanhui;GUO Youmin;LI Jianlong(Department of Radiology Affiliated Hospital of Yan'an Universit Yan'an,Shaanxi Province 716000,China;Department of Radiology,the First Affiliated Hospital of Xi'an Jiaotong Universitycm 710061,China)

机构地区：[1]延安大学附属医院医学影像科,陕西延安716000 [2]西安交通大学第一附属医院影像科,陕西西安710061

出　　处：《实用放射学杂志》2020年第11期1845-1848,共4页Journal of Practical Radiology

摘　　要：目的探讨Karl迭代重建的超低剂量CT对全肺平均肺密度(MLD)和全肺容积(LV)定量分析的影响.方法前瞻性选取固定每年健康体检者58例,首次行胸部低剂量(120 kV,40 mAs)CT扫描,次年行超低剂量(120 kV,3 mAs)胸部CT扫描.低剂量采用滤波反投影(FBP)重建,超低剂量采用FBP和Karl迭代重建.在肺动脉分叉层面测量肺动脉主干图像噪声标准差(SD值),将CT数据导入"数字肺"检测及分析平台,定量测量全肺MLD和LV.采用5分法主观评价不同重建图像质量.采用单因素方差分析或Wilcoxon符号秩检验比较各组MLD、LV、SD及视觉评分间的差异,P<0.05为差异有统计学意义.结果低剂量组和超低剂量组有效辐射剂量(ED)分别为(1.70±0.13)mSv和(0.13±0.01)mSv,超低剂量组ED下降约92%.与低剂量组相比,超低剂量组CT对全肺LV定量结果的准确性无影响.和低剂量FBP组相比,超低剂量FBP组的MLD高于低剂量FBP组,超低剂量Karl迭代重建组的MLD趋于低剂量组.噪声方面超低剂量FBP组最高,和其相比超低剂量Karl迭代重建组SD减少约27%.低剂量FBP组视觉评分最高,其次为超低剂量Karl迭代重建组,超低剂量FBP组最低.结论超低剂量CT在降低有效辐射剂量约92%的前提下,采用Karl迭代重建能降低图像噪声提高图像质量,满足CT对全肺LV的定量分析,能提高全肺平均肺密度检测的准确性.Objective To investigate the effect of ultra-low-dose CT reconstructed by Karl iteration on the quantitative analysis of mean lung density(MLD)and lung volume(LV).Methods Fifty-eight healthy subjects who were fixed annually were prospectively collected.Low-dose(120 kV,40 mAs)CT scan of the chest was performed for the first time,and the ultra-low-dose(120 kV,3 mAs)CT scan of the chest was performed for the following year.Low-dose reconstruction was performed using filtered back projection(FBP),and ultra-low-dose reconstruction was performed using FBP and Karl iteration.Standard deviation(SD value)of the image noise at the main pulmonary artery was measured at the level of pulmonary artery bifurcation.CT data was imported into the"digital lung"detection and analysis platform to quantitatively measure the whole lung MLD and LV.5-point method was used to subjectively evaluate the quality of different reconstructed images.Univariate analysis of variance or Wilcoxon sign rank test was used to compare the differences between MLD,LV,SD and visual scores in each group.P value which was less than 0.05 was regarded as statistically significant.Results The effective dose(ED)of the low-dose group and ultra-low-dose group was(1.70±0.13)mSv and(0.13±0.01)mSv,respectively.The ultra-low-dose group ED decreased by about 92%,Compared with low-dose group,ultra-low-dose CT group had no effect on the accuracy of quantification of lung LV.Compared with the low-dose FBP group,the MLD of the ultra-low-dose FBP group was higher than that of the low-dose FBP group,and the MLD of the ultra-low-dose Karl iterative reconstruction group tends to the low-dose group.In terms of image noise,the ultra-low-dose FBP group had the highest SD,and SD was reduced by about 27%in the ultra-low-dose Karl iterative reconstruction group.The low-dose FBP group had the highest visual score,followed by the ultra-low-dose Karl iterative reconstruction group,and the ultra-low-dose FBP group had the lowest score.Conclusion Under the premise of reducing the effec

关 键 词：滤波反投影 迭代重建 平均肺密度 肺容积 噪声 超低剂量 

分 类 号：R814.42[医药卫生—影像医学与核医学] R816.41[医药卫生—放射医学]