双源双能量CT单能谱技术在肺动脉及下肢静脉联合成像中的应用价值

Impact of Dual Energy Virtual Monochromatic Imaging on Image Quality and Diagnostic Accuracy at the Combined Computed Tomography Angiography of Pulmonary and Lower Extremity Vein

目的研究双源双能量噪声优化的虚拟单能谱成像技术(VMI+)对下肢CT静脉成像(CTV)和肺动脉成像(CTPA)联合扫描中下肢静脉图像质量的影响,且以超声为参考标准评估CTV对下肢静脉血栓的诊断性能。方法回顾性分析53例临床怀疑肺栓塞(PE)进行CTPA-CTV联合检查的患者的图像,利用双源双能量单能谱成像技术分别重建下肢静脉在40、50、60keV的三组虚拟单能谱图像(VMI+40 ke V、VMI+50keV、VMI+60keV),以及线性融合图像(融合系数0.5)。记录线性融合图像和三组单能谱图像的衰减值(CT值)、对比噪声比(CNR)及信号噪声比(SNR),以三分制对图像进行主观评分。以超声(US)对血栓的诊断结果为参考标准,计算下肢CTV诊断下肢深静脉血栓(DVT)的特异性、敏感性、阳性预测值及阴性预测值;使用Kappa检验分析两种方法对DVT显示的一致性。结果单能谱图像和线性融合图像的CT值及CNR值存在统计学差异(H=116.455,P=0.000及H=22.362,P=0.000),其中VMI+40 keV的CT值(204.63±56.40HU)及CNR值(7.72±4.72)最高。四组图像的主观评分存在统计学差异(H=14.097,P=0.003),其中VMI+40 keV的评分(2.62±0.49)最高。在37位同时进行CTV及US检查的患者中,CTV对DVT的诊断敏感性、特异性、阳性预测值及阴性预测值分别为84.2%(16/19)、77.8%(14/18)、80.0%(16/20)及82.4%(14/17)。结论双源双能量单能谱成像技术显著提高肺动脉及下肢静脉联合成像中下肢静脉的图像质量,且具有较高的血栓诊断性能。

Objective To investigate the impact of dual-energy noise-optimized virtual monoenergetic imaging algorithm(VMI+)on image quality and diagnostic performance of lower-extremity indirect computed tomography venography(CTV)combined with pulmonary angiography(CTPA)for venous thromboembolism,compared with Ultrasound.Methods 53 patients who were clinically suspected pulmonary embolism(PE)and underwent CTPA-CTV examination on a third-generation dual-source CT system were retrospectively enrolled and analyzed in the study.Images were reconstructed with standard linear blending(F0.5),VMI+at 40 keV(VMI+40 keV),50 keV(VMI+50 keV)and 60 keV(VMI+60 ke V).Vascular attenuation values and image noise were measured;contrast to noise ratio(CNR)and signal to noise ratio(SNR)of VMI+images and linear blending images were calculated.Threepoint scales were used to evaluate the subjective image quality.In a subgroup of 37 patients who additionally underwent US check,the specificity,sensitivity,positive predictive value,and negative predictive value of CTV for the detection of deep venous thrombosis(DVT)were calculated.Kappa test was used to analyze the consistency of the two methods for DVT.Results The value of attenuation and CNR were highest in the VMI+40 keV group(204.63±56.40 HU and 7.72±4.72,respectively).The subjective assessment was highest in the VMI+40 keV(2.62±0.49),superior to F0.5(2.30±0.46,P<0.05).Of the 37 patients who underwent CT and US examination,the sensitivity,specificity,positive predictive value and negative predictive value of CTV in the diagnosis of deep venous thrombosis were 84.2%(16/19),77.8%(14/18),80.0%(16/20),and 82.4%(14/17)respectively.Conclusion Compared with the linear blending image,low-kiloelectron volt VMI+can improve the image quality of CTV and diagnostic accuracy for deep venous thrombosis.