全模型迭代重建算法(IMR)在“双低”冠状动脉CTA成像中应用价值的探讨

摘要

目的:探讨全模型迭代重建(Iterative model reconstruction,IMR)技术在256层螺旋CT“双低”冠状动脉CTA成像中的应用价值.方法:100例拟行冠状动脉CTA检查者按照同期随机原则分为2组,每组50例.均采用前置门控扫描.A组(对照组)采用120 kV、管电流根据噪声指数自动调节DRI=15 (Dose Right Index)及iDose4重建,B组(低剂量组),采用100 kV、DRI=11及IMR重建.A、B组对比剂注射方案分别为0.8 mL/kg体质量和0.6 ml/kg体质量,对比剂注射时间均为12s.采用独立样本t检验比较两组客观评价指标(噪声、信噪比及对比噪声比)、辐射剂量(CT剂量指数、CT剂量长度乘积、有效辐射剂量)及碘负荷的差异.采用Mann-Whitney U秩和检验比较两组主观图像(管腔对比度、边缘锐利度、主观噪声及图像总体质量)质量评分的差异.结果:B组噪声为(23.4±2.8) HU,低于A组(34.3±4.3) HU,差异有统计学意义(t=14.977,P=0.000).B组信噪比为(16.3±2.2),高于A组(11.9±1.9),差异有统计学意义(t=-10.713,P=0.000).B组对比噪声比为(21.1±2.7),高于A组(15.1±2.3),差异有统计学意义(t=-12.083,P=0.000).B组主观噪声与整体质量优于A组,差异有统计学意义(U=951.000、934.000,P=0.008、0.005),两组对比度与锐利度差异无统计学意义(U=1 233.000、1 233.500,P=0.884、0.888).B组有效辐射剂量为(0.94±0.23) mSv,较A组剂量(1.53±0.32)mSv降低约39％;B组碘负荷为(15.1±2.4)g,较A组(20.1±2.5)g降低约25％.结论:IMR较混合迭代重建(iDose4)可以进一步降低噪声,在剂量降低39％且碘负荷降低25％的条件下,图像质量仍可满足诊断需求.%Objective:To investigate the clinical value of IMR technique in 256 slice prospective coronary CTA with low dose and low Iodine load.Methods:One hundred patients who would perform coronary CTA were enrolled and the patients were divided into two groups randomly.All the patients were performed prospective CTA.Patients in group A were performed CTA by using normal tube voltage(120 kV),dose right index(DRI)=15 was considered to be currents and iDose4 reconstruction.Patients in group B were performed CTA by using low tube voltage (100kV),DRI=11 was considered to be currents and IMR reconstruction.The injection protocols of group A and group B were 0.8 mL/kg and 0.6 mL/kg,respectively.The injection time was 12 seconds.Quantitative measurements of CT value,image noise,signal-to-noise (SNR) and contrast-to-noise (CNR) were measured in each group.t-test was used for comparisons of objective evaluation,indices(noise,SNR,CNR),radiation dose(CTDIvol,DLP,ED) and Iodine load between the two groups.Mann-Whitney U test was used for comparisons of subjective evaluation of IQ between the two groups.A level of P＜0.05 was considered statistically significant.Results:There were significant differences in SD between group A(34.3±4.3)HU and group B(23.4±2.8)HU(t=14.977,P=0.000).There were significant differences in SNR between group A (11.9±1.9) and group B (16.3±2.2)(t=-10.713,P=0.000).There were significant differences in CNR between group A(15.1±2.3) and group B(21.1±2.7)(t=-12.083,P=0.000).Group B showed the better objective data.Compared with group A,there were significant differences in subjective noise and acceptability of images (U=951.000,934.000,P=0.008,0.005).There were no significant differences in image contrast and image sharpness (U=1 233.000,1 233.500,P=0.884,0.888).The effective dose(ED) of group B (0.94±0.23)mSv was 39％ lower than that of group A (1.53±0.32)mSv.The Iodine load of group B (15.1±2.4)g was 25％ lower than that of group A (20.1±2.5)g.Conclusion:Compared with iDose4,using 256 MDCT,iterative model reconstruction technique can provide 39％ ED and 25％ Iodine load reduction,while the images could still satisfy the requirement of diagnosis.