胰腺低剂量容积灌注CT的相关研究

发布时间：2018-07-21 21:14

【摘要】：目的:评估低剂量胰腺灌注CT灌注参数测量的可靠性及图像质量。材料和方法:本研究获得了伦理委员会的批准,患者检查前均签署书面的知情同意书。最终入组100例因怀疑胰腺占位行低剂量胰腺灌注检查患者。扫描条件采用以下5组组合形式:A组,管电压80Kv,管电流120mAs(体重≥80Kg);B组,管电压80Kv,管电流100mAs(体重≥70Kg);C组,管电压70Kv,管电流189mAs(体重≥60Kg);D组,管电压70Kv,管电流130mAs(体重≥50Kg);E组,管电压80Kv,管电流55mAs(体重50Kg)。由1名经验丰富的放射科医生记录剂量参数CDTI和DLP,并用去卷积法测量灌注参数,包括血流量(blood flow,BF),血容量(blood volume,BV)和渗透性(permeability,PMB)。选择灌注的平扫、动脉、门脉期图像,评估灌注图像的图像质量,包括主观图像质量评分及客观图像质量参数(图像背景噪声、信噪比、对比噪声比)。单因素ANOVA分析各组间图像质量及灌注参数比较。结果:各低剂量组较我院常规组(80Kv,150mAs)相比辐射剂量有不同程度的降低,A-D低剂量组图像质量相当(P0.05),E组图像质量高于其它低剂量组(P0.05)。低剂量组间所测量的正常胰腺实质的灌注参数BF、BV、PMB无显著性差异,P值均大于0.05。结论:低剂量胰腺容积灌注在保证图像质量的前提下,可以有效的降低辐射剂量,其所测得的正常胰腺实质的灌注参数真实可靠。目的:评估低剂量灌注CT图像数据用于三维重建的图像质量。材料和方法:研究对象为论文第一部分收集的低剂量胰腺灌注扫描患者,根据灌注形成的时间密度曲线,选择腹主动脉、门脉主干强化最佳时期行三维重建图像,三维重建包括表面漫游技术(volume rendering technique,VRT)和最大密度投影(Maximum intensity projection,MIP),灌注及多期增强的动脉期行VRT重建,灌注及多期增强的门脉期行MIP重建,由2名放射科医师对其图像进行主观和客观评价(背景噪声、信噪比),采用配对样本的非参数检验比较灌注数据与多期增强三维重建的图像质量。结果:灌注的VRT及MIP图像,其血管光滑程度、血管分支显示情况,优于常规多期增强,但其图像的噪声高于常规增强,但都无统计学意义。两者VRT及MIP的总体图像质量无明显差别。结论:根据灌注的时间密度曲线选择的灌注图像可以代替常规多期增强图像行血管的三维重建。目的:探讨灌注CT的功能灌注参数对胰腺神经内分泌肿瘤进行术前病理分级(G1、G2、G3)。材料和方法:回顾性分析2015年12月至2017年2月北京协和医院怀疑为胰腺神经内分泌肿瘤而行灌注CT检查患者77例,其中在我院行手术切除并经病理证实为胰腺神经内分泌肿瘤(pancreatic neuroendocrine tumors,pNETs)患者 37例,其中G1级19例,G2级17例,G3级1例。将G1-G3级肿瘤分为A、B两组,A组为G1级肿瘤,B组G2/3级肿瘤。所有患者均采用标准化胰腺灌注扫描方案。由1名经验丰富的放射科医师在西门子后处理工作站上测量并记录肿瘤及正常胰腺实质的灌注参数:血流量(blood flow,BF)、血容量(blood volume,BV)、渗透性(permeability,PMB),测量结果用均数±标准差表示。用Spearman相关分析分析相关性,用独立样本T检验分析A、B两组灌注参数的差异性。并绘制受试者工作特征曲线(Receiver operating characteristic curve,R0C)。结果:无论G1还是G2/3级肿瘤BF值均与正常胰腺实质存在显著差异(p=0.000,p=0.018),PMB均无统计学差异(p=0.569,p=0.174),而BV在G1中存在显著性差异(p=0.000),在G2/3级中无明显差异(p=0.763)。Spearman分析显示,A、B两组灌注参数BF、PMB与肿瘤病理分级之间无相关性(r=-0.230,P=0.177;r=-0.134,P=0.437),BV 与 G1-G3 之间存在负相关性(r=-0.444,P=0.007),BV值越低其病理分级越高。独立样本T检验显示A、B两组患者灌注参数BF、PMB无显著性差异(P=0.155,P=0.646),而BV存在明显差异(P=0.023)。B组患者的BV值明显低于A组。R0C曲线显示BV诊断肿瘤分级时其曲线下面积为0.756。BV判断肿瘤为G2(G3)的最佳截点值为23.0,此时其敏感性为72.0%,特异性为77.8%。结论:胰腺神经内分泌肿瘤病理分级不同,其灌注参数BV间存在显著性差异,BV值与肿瘤分级之间存在负相关性,BV值可以术前鉴别胰腺神经内分泌肿瘤的病理分级,BV值越低其病理分级越高。
[Abstract]:Objective: To evaluate the reliability and image quality of low dose pancreatic perfusion CT perfusion parameters. Materials and methods: This study was approved by the ethics committee and signed written informed consent before the examination. 100 patients with low dose pancreatic perfusion suspected to be suspected of pancreatic space occupancy were enrolled in the study. The following 5 groups were used. Combined form: A group, tube voltage 80Kv, tube current 120mAs (weight > 80Kg); B group, tube voltage 80Kv, tube current 100mAs (weight > 70Kg); C group, tube voltage 70Kv, tube current 189mAs (weight > 60Kg); group, tube voltage, tube current (weight >); tube current, tube voltage, tube current (weight). Recorded by 1 experienced radiologists. Dose parameters CDTI and DLP, and use deconvolution method to measure perfusion parameters, including blood flow (blood flow, BF), blood volume (blood volume, BV) and osmosis (permeability, PMB). Choose the plain scan, arterial, portal phase images, evaluate the image quality of the perfusion image, including the subjective image quality score and the objective image quality parameters (image background). Noise, signal to noise ratio, contrast to noise ratio). Comparison of the image quality and perfusion parameters of the single factor ANOVA analysis. Results: compared with the conventional group (80Kv, 150mAs), the radiation dose of each group was lower than that of the conventional group (80Kv, 150mAs). The image quality of the low dose group of A-D was equal (P0.05), and the image quality of the group E was higher than that of the other low dose group (P0.05). There was no significant difference in the perfusion parameters of the normal pancreatic substance BF, BV and PMB, and the P values were all greater than the 0.05. conclusion: low dose pancreatic volume perfusion could effectively reduce the radiation dose on the premise of ensuring the quality of the image. The perfusion parameters of the normal pancreatic substance were true and reliable. Objective: To evaluate the low dose perfusion CT image data. The image quality of three-dimensional reconstruction. Materials and methods: the research object was the low dose pancreatic perfusion scan patients collected in the first part of the paper. According to the time density curve of the perfusion formation, the abdominal aorta was selected and the three dimensional reconstruction image was performed at the best time of the portal backbone strengthening, and the 3D reconstruction including volume rendering technique, V RT) and the maximum density projection (Maximum intensity projection, MIP), perfusion and multiphase enhanced arterial phase, VRT reconstruction, perfusion and multiphase portal phase of MIP reconstruction, subjective and objective evaluation of the images by 2 radiologists (background noise, signal to noise ratio), and non parametric test of paired samples to compare perfusion data and more Results: the image quality of the three dimensional reconstruction was enhanced. Results: the VRT and MIP images of the perfusion were better than the conventional multiphase enhancement, but the image noise was higher than the conventional enhancement, but there was no significant difference in the image quality. There was no significant difference in the overall image quality of both VRT and MIP. Conclusion: according to the time density curve of perfusion. The selected perfusion image can replace the three-dimensional reconstruction of the routine multiphase enhanced image. Objective: To investigate the preoperative pathological classification of pancreatic neuroendocrine tumor (G1, G2, G3) by perfusion CT perfusion parameters. Materials and methods: retrospective analysis from December 2015 to February 2017 in Peking Union Medical College Hospital was suspected as the pancreatic neuroendocrinology. 77 patients were treated with CT, including 37 cases of pancreatic neuroendocrine tumor (pancreatic neuroendocrine tumors, pNETs) in our hospital, of which 19 cases were G1 grade, 17 cases of G2 grade, 1 cases of G3 grade, and G1-G3 grade tumors were divided into A, B two, A group was grade tumor, and all patients were labeled. Quasi pancreatic perfusion scan. 1 experienced radiologists measured and recorded perfusion parameters of tumor and normal pancreatic parenchyma on a SIEMENS post processing workstation: blood flow (blood flow, BF), blood volume (blood volume, BV), permeability (permeability, PMB), and the measured results were expressed with mean standard deviation. Analysis of correlation and analysis of the difference in perfusion parameters between A and B two groups by independent sample T test. And draw the work characteristic curve of the subjects (Receiver operating characteristic curve, R0C). 174) and there was a significant difference in G1 (p=0.000), and there was no significant difference in the G2/3 level (p=0.763).Spearman analysis showed that there was no correlation between the perfusion parameters of A, B two, PMB and tumor pathological grading (r=-0.230, P=0.177; p=0.000), and the lower the value, the higher the pathological grade. The independent sample T test showed that the perfusion parameters of A, B two were BF, PMB had no significant difference (P=0.155, P=0.646), but there was a significant difference in BV (P=0.023), the BV value of the patients in the.B group was significantly lower than that of the A group. .0%, the specificity is 77.8%. conclusion: the pathological classification of pancreatic neuroendocrine tumor is different, there is a significant difference between the perfusion parameters BV and the negative correlation between the BV value and the tumor classification. The BV value can identify the pathological grade of the pancreatic neuroendocrine tumor before operation, and the lower the BV value, the higher the pathological grade.
【学位授予单位】：北京协和医学院
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R816.5

【相似文献】