探究靶重建结合能谱CT最佳单能量在门静脉成像中的应用价值

To Explore the Application Value of Target Reconstruction Combined with Spectral CT Optimal Single Energy in Portal vein Imaging

目的探究靶重建结合能谱CT最佳单能量在CT门静脉成像(CTPV)中的应用价值。方法回顾性搜集2022年4月至2022年8月行上腹部CT增强扫描的患者40例。采用宝石能谱成像(GSI)模式进行扫描,管电压80/140 kVp瞬时切换,扫描完成后,对原始数据进行图像重组,常规组(A组):重组70 keV、40%自适应统计迭代重组算法(ASIR-V)结合常规大视野(DFOV)图像(包含腹部四周皮缘);实验组(B组):60~70 keV(间隔2 keV)、80%ASIR-V结合靶重组(DFOV仅包括整个肝脏组织)共6个亚组图像,记为B1~B6组。记录A、B两组的DFOV大小。将图像传至GE AW 4.7后处理工作站,重组门静脉容积再现(VR)、最大密度投影(MIP)及肝脏三维可视化图像。在门静脉期轴位图像上测量门静脉、肝脏的CT值、标准差(SD)值,计算门静脉对比噪声比(CNR)。由两名观察者采用五分法评估各组图像质量及门静脉显示情况,并行Kappa一致性检验。结果A、B两组的DFOV大小分别为(49.64±6.74)cm、(27.19±4.36)cm(t=23.49,P<0.05)。两名观察者对各组图像的评分一致性较好(Kappa值0.764~0.936,P<0.05)。与A组比较,B组图像主观评分及门静脉分支显示评分均高于A组(P<0.05);B1组门静脉CT值与A组无统计学差异(P=0.430),B2~B6组CT值均高于A组(P<0.05);B1、B2组SD值低于A组(P<0.05),B3组SD值与A组无统计学差异(P=0.517),B4~B6组SD高于A组(P<0.05),B1~B6组CNR均高于A组(P<0.05)。组内比较:随着keV的降低,B1~B6组门静脉CT值、SD值逐渐升高(P<0.05),CNR逐渐升高(P<0.05);B1~B6组图像主观评分及门静脉分支的清晰显示评分呈先上升后降低的趋势(P<0.05),B3组(66 keV)得分最高。结论采用靶重建结合66 keV能够显著提高肝内门静脉的显示,为临床诊疗提供更精确的影像信息。

Objective To explore the application value of target reconstruction combined with optimal monochromatic energy level in spectral CT portal vein imaging(CTPV).Methods 40 patients for upper abdominal DECT angiography were retrospectivelycollected from April 2022 to August 2022.The original data were reconstructed and divided into two groups according to DFOV.Group A wasreconstructed at 70 keV and 40%ASIR-V,combined with conventional DFOV image,covering the abdominal skin margin.Group B was reconstructedfrom 60 to 70 keV(2 keV interval)and 80%ASIR-V,combined with target reconstruction,including the whole liver only,and six subgroups(B1-B6)were obtained.The size of DFOV was recorded in group A and B.The images were transferred to GE AW 4.7 workstation to reconstruct portal vein volume reproduction(VR),maximum density projection(MIP)and liver 3D visualization images.The CT and SD values of portal vein and liver were measured,and the CNR of portal vein was calculated.Two observers used the five-point method to evaluatethe image quality and the display ofportal vein.Kappa test was used to asses the inter-observer consistency.Results The DFOV size of group A and B was(49.64±6.74)cm and(27.19±4.36)cm,respectively(t=23.49,P<0.05).Two observers had good consistency in theevaluation of image quality and the display of portal vein(Kappa value=0.764-0.936,P<0.05).The subjective score and portal vein display score in group B were higher than those in group A(P<0.05).There was no significant difference in the CT value of portal vein between group B1 and group A(P=0.430),while the CT value of group B2-B6 was higherthan that of group A(P<0.05).Compared with group A,the SD of group B1 to B2 was lower but the SD value of group B4 to B6 was higher(P<0.05).No significant difference was found between group B3 and group A(P=0.517).The CNR of groups B1 to B6 was higher than that of group A(P<0.05).Intra-group comparison:with the decrease of keV,CT and SD values of portal vein in B1-B6 groups gradually increased(P<0.05),and CNR gradually increased as well(P<0.05).The subjective score for the portal vein branches in group B1 to B6 increased with the decrease of keV initially then gradually decreased(P<0.05),and the score of group B3(66 keV)was the highest.Conclusion Target reconstruction combined with 66 keV can significantly improve the display of intrahepatic portal vein and provide more accurate image information for clinical diagnosis and treatment.