宽体探测器单源双能CT肝脏图像质量评价

摘要

目的:与单源双能HD750CT机进行对比, 评价宽体探测器单源双能RevolutionCT机肝脏图像质量.方法:回顾性收集2017年10月至11月应用RevolutionCT机进行腹部增强检查 (包含上腹) 的患者, 为A组;以及同时期在HD 750CT机进行腹部增强检查 (包含上腹) 的患者, 为B组.A组37例, B组39例.扫描条件设置, A组:ASIR-V40％, 探测器宽度80mm, 螺距0.992:1;B组:ASIR40％, 探测器宽度40mm, 螺距1.375:1.扫描范围包括整个肝脏, 进行平扫及增强三期扫描.将5mm平扫mono图像 (70keV) 及门静脉期单能量图像 (40keV、50keV、70keV) 传至AW4.6工作站, 记录容积CT剂量指数 (CTDIvol).由两位医师 (分别工作5年, 3年) 共同对两组图像进行4分制主观评价 (4分好, 1分差).两位医师采用双盲法, 使用AW4.6工作站对5mm平扫mono图像 (40keV、50keV、70keV) 及门静脉期图像 (40keV、50keV、70keV) 分别各放置三个感兴趣区 (regionofinterest, ROI) 进行测量, 测量数据包括平扫及门静脉期图像中第一肝门层面肝右叶的CT值和SD值, 平扫图像中第一肝门层面右侧背部皮下脂肪的CT值和SD值, 门静脉期图像中第一肝门层面右侧竖脊肌的CT值和SD值.并计算肝脏的CNR及SNR.所测量数据分别取均值.采用组内相关系数 (ICC) 比较两位医师所测量数据的一致性, 选择高年资医师所测数据进行后续统计学分析.应用Mann-WhitneyU比较两组间管电流, 主观评分及平扫图像肝脏CNR的差异性, 应用独立样本t检验比较两组间年龄、腹围、CTDIvol、平扫图像肝脏SD值、SNR及门静脉期 (40keV、50keV、70keV) 图像肝脏SD值、CNR、SNR的差异性.结果:两位医师测量数据一致性良好 (ICC均>0.8).A组平扫图像 (40keV、50keV) 肝脏SD值低于B组, 肝脏CNR和SNR高于B组, 差异具有统计学意义 (P<0.05).A组平扫图像 (70keV) 肝脏CNR及SNR高于B组 (P<0.05) , 两组平扫图像 (70keV) 肝脏SD值无统计学差异 (P>0.05).A组门静脉期图像 (40keV、50keV、70keV) 肝脏SD值低于B组, CNR和SNR高于B组 (P<0.05).A组图像主观评分均高于B组, 其中平扫图像 (70keV) 主观评分差异无统计学意义 (P>0.05) , 平扫图像 (40keV、50keV) 及门静脉期图像 (40keV、50keV、70keV) 两组主观评分差异均有统计学意义 (P<0.05).两组管电流分别为 (369±30) mA和 (367±25) m A, 差异无统计学意义 (P>0.05).A组CTDIvol低于B组 (P<0.05) , 分别为 (10.60±1.22) mGy和 (11.04±0.52) mGy.结论:宽体探测器RevolutionCT机与HD750CT机相比, 能够在低辐射剂量情况下保证更低的噪声 (SD值) , 图像均匀度更好, 图像质量更好 (主观评分更高, SNR越高) , 尤其对于门静脉期图像质量的提高更为显著.%Purpose: To explore the image quality of liver enhanced CT imaging with wide-body detector single source dual energy CT, and compared with that with HD750 CT. Methods: Patients who were scanned with Revolution CT scanner were treated as group A, and those with HD750 CT scanner were treated as group B.They were all enrolled in this retrospective study in October and November 2017. The group entry standard was: a.scanning mode was energy spectrum scanning mode; b. no occupying lesion in the liver, no fatty liver; c. abdominal circumference was between 80 cm-85 cm. There were 37 patients in group A and 39 patients in group B for the final group. The spectral imaging acquisition protocol with Revolution CT was: ASIR-V 40％, detector width 80 mm, pitch 0.992:1; and that with HD 750 CT was: ASIR40％, detector width 40 mm, pitch 1.375:1. Pre-contrast and three enhanced phases were obtained covering the whole liver. The 5 mm plain scan mono image (70 keV) and the portal phase single energy image (40 keV, 50 keV, 70 keV) were transferred to the AW 4.6 workstation. The volume CT dose index (CTDIvol) was recorded. Two doctors (with 5 years and 3 years' work experience respectively) made a subjective evaluation of the two groups of images on the basis of 4 points. CT values, SD values in 40 keV, 50 keV and 70 keV of the right lobe of the liver, subcutaneous fat and erector spinae were measured. CNR and SNR were computed. The difference of the two groups was analyzed with Mann-Whitney U or Independent sample t test.Results: The data measured by two doctors were in good agreement (ICC>0.8). There were significant differences in SD values of liver, CNR and SNR between the two groups in 40 keV and 50 keV for pre-contrast images. There were significant differences in CNR and SNR between the two groups whereas there was no difference in SD values of liver in 70 keV for pre-contrast images. There were significant differences in SD values of liver, CNR and SNR between the two groups in 40 keV, 50 keV and 70 keV for portal vein phase images. The subjective score of group A was higher than that of group B. There were significant differences in the subjective score between the two groups whereas there was no difference in 70 keV for pre-contrast images. The current of the two groups was (369±30) mA and (367±25) mA, respectively, and there was no significant difference between them (P<0.05). The CTDIvol of group A was lower than that of group B (10.60±1.22 mGy and 11.04±0.52 mGy respectively). Conclusion: Compared with HD 750 CT, the wide body detector revolution CT machine can guarantee lower noise (SD value) and better image uniformity under low radiation dose. The higher the subjective score, the higher the image quality, especially in the portal phase.