摘要目的 探讨256层CT混合迭代重建算法对改善颈动脉入颅段CTA图像质量及辐射剂量的价值.方法 对60例患者行256层CT头颈CTA联合成像,其中30例行常规剂量扫描,分别采用滤波反投影(FBP)重建生成图像(A组)及迭代iDose4重建生成图像(B组),30例行低剂量扫描分别采用FBP重建(C组)及iDose4重建(D组),并对四组图像的噪声、信噪比(SNR)、对比噪声比(CNR)、图像质量主观评分及有效辐射剂量进行评价.采用单因素方差分析比较各组间客观图像质量评价指标,采用等级资料的多个独立样本Kruskal-Wallis检验比较各组间主观图像质量评分,采用配对t检验比较常规剂量组和低剂量组的有效辐射剂量.结果 4组图像噪声分别为8.21 ±0.88、7.31±0.33、11.17 ±2.02和6.50 ±0.49,SNR分别为43.21±4.49、51.83±3.64、42.88 ±9.19和53.47±4.88,CNR分别为37.88±4.02、61.21±6.31、36.63±8.20和62.99 ±5.90,差异均有统计学意义(F值分别为112.786、97.041和86.098,P值均＜0.01).除B组和D组的图像噪声差异无统计学意义外,其余两组差异具有统计学意义(P值均＜0.01)；除A组和C组、B组和D组的SNR和CNR差异无统计学意义外,其余两组差异具有统计学意义(P值均＜0.01).图像质量评分分别为1、2、3分者,A组分别为2、13和15例,B组分别为0、7和23例,C组分别为5、15和10例,D组分别为0、5和25例,各组图像质量评分差异有统计学意义(H值=22.575,P值＜0.01).常规剂量组有效辐射剂量为(2.31 ±0.13) mSv,低剂量组为(0.84±0.04) mSv,差异有统计学意义(t=60.682,P＜0.05).结论 迭代重建算法在颈动脉入颅段CTA成像中,在辐射剂量降低50％以上的情况下仍能获得优良的诊断图像.

abstractsObjective To explore the role of 256 CT iterative reconstruction in improving image quality and reducing radiation dose of carotid artery into cranial segment.Methods Sixty patients underwent head and neck CTA on a 256 CT scanner.Conventional dose scanning was performed in 30 patients using a filtered back projection (FBP) reconstruction (group A) and iterative iDose4 reconstruction (group B).Low dose scanning was performed in the other 30 patients using FBP reconstruction (group C) and iDose4 reconstruction (group D).The noise,SNR,CNR,score of image quality and effective radiation dose were evaluated in four groups.One-way ANOVA was used to analyze the image quality index between groups.Independent sample of Kruskal-Wallis test of ranked data was used to compare image quality score between groups.Paired t test was used to compare the effective radiation doses between the low dose group and conventional dose group.Results The image noise of four groups were 8.21 ±0.88,7.31 ± 0.33,11.17 ± 2.02 and 6.50 ± 0.49 respectively.SNR were 43.21 ± 4.49,5 1.83 ± 3.64,42.88 ± 9.19 and 53.47 ± 4.88,respectively.CNR were 37.88 ± 4.02,61.21 ± 6.31,36.63 ± 8.20 and 62.99 ±5.90,respectively.There were statistic differences (F =112.786,97.041 and 86.098,P ＜0.01).The differences of image noise between group A and B had no statistic significance,which was statistic different between group A and C (P ＜ 0.01).Except that the differences of SNR and CNR between group A and C,B and D had no statistic significance,the differences between other two groups had statistic significance (P ＜ 0.01).In case of image quality score of 1,2 and 3,there were 2,13 and 15 patients in group A; 0,7 and 23 patients in group B; 5,15 and 10 patients in group C; and 0,5 and 25 patients in group D.There was statistic differences of image quality score between each group (H =22.575,P ＜0.01).The effective radiation dose was (2.31 ±0.13) mSv in conventional dose group and (0.84 ±0.04) mSv in low dose group.There was statistic difference between the two groups (t =60.682,P ＜ 0.05).Conclusion Compared with conventional dose,iDose4 iterative reconstruction algorithms can obtain excellent images of CTA for carotid artery into the cranial segment with more than 50％ radiation dose decrease.