动态对比增强MRI定量参数直方图分析预测胃癌病理分级的价值  

The value of dynamic contrast-enhanced MRI quantitative parameters using histogram analysis in predicting the pathological grades of gastric cancer

作　　者：张霄 张宏凯 路双 贾争艳 赵玉洲 黎海亮 曲金荣 ZHANG Xiao;ZHANG Hongkai;LU Shuang;JIA Zhengyan;ZHAO Yuzhou;LI Hailiang;QU Jinrong(Department of Radiology,the Affiliated Cancer Hospital of Zhengzhou University,Zhengzhou 450008,China;Department of General Surgery,the Affiliated Cancer Hospital of Zhengzhou University,Zhengzhou 450008,China)

机构地区：[1]郑州大学附属肿瘤医院放射科,河南郑州450008 [2]郑州大学附属肿瘤医院普外科,河南郑州450008

出　　处：《实用放射学杂志》2022年第1期64-67,84,共5页Journal of Practical Radiology

基　　金：河南省中青年卫生健康科技创新领军人才培养项目(YXKC2020011)。

摘　　要：目的 探讨动态对比增强磁共振成像(DCE-MRI)定量参数直方图分析预测胃癌分化程度的价值.方法 回顾性分析病理证实为胃癌患者的资料,测量术前DCE-MRI全域直方图参数,包括容积转移常数(Ktrans)、速率常数(Kep)及血管外细胞外间隙容积比(Ve)的平均值、第10百分位数、第90百分位数、偏度、峰度和熵.采用Kruskal-Walls H检验及Mann-WhitneyU检验比较参数差异.利用受试者工作特征(ROC)曲线评价有统计学意义的参数的诊断效能.结果 低分化组38例、中分化组38例、高分化组6例.低分化组Ktrans及Kep的平均值、第10百分位数、第90百分位数、熵和Ve第10百分位数高于中、高分化组,Kep偏度、峰度低于中、高分化组,差异有统计学意义(P<0.05).Ktrans第10百分位数鉴别低分化组的效能最高,曲线下面积(AUC)为0.803,敏感度为68.42%,特异度为86.36%.结论 DCE-MRI定量参数直方图分析有助于术前评估胃癌分化程度,Ktrans第10百分位数诊断效能最高.Objective To explore the value of histogram analysis of dynamic contrast-enhanced magnetic resonance imaging(DCE-MRI)in predicting pathological grades in gastric cancer.Methods The data of patients with gastric cancer diagnosed by pathological findings were retrospectively collected.Whole tumor histogram quantitative parameters of DCE-MRI including mean,10th percentile,90th percentile,skewness,kurtosis,entropy of K^(trans),K_(ep)and V_(e)were measured.Kruskal-Walls H test and Mann-Whitney U test were used to compare the parameters.Receiver operating characteristic(ROC)curve was used to evaluate the diagnostic efficacy of the statistically significant parameters.Results There were 38 cases in the poorly differentiated group,38 cases in the moderately differentiated group and 6 cases in the well differentiated group.The mean,10th percentile,90th percentile,entropy of K^(trans) and K_(ep) and 10th percentile of Ve in the poorly differentiated group were higher than those in the moderately and well differentiated groups.The skewness and kurtosis of Kep in the poorly differentiated group were lower than those in the moderately and well differentiated groups.The difference was statistically significant(P<0.05).The 10th percentile of K,rans was the most effective parameter in identifying poorly differentiated group and the area under the curve(AUC)was 0.803.Its sensitivity and specificity were 68.42%and 86.36%,respectively.Conclusion Histogram analysis of DCE-MRI quantitative perfusion parameters contributes to evaluate preoperative pathological grades of gastric cancer and the 10th percentile of K^(trans) is the most effective parameter.

关 键 词：胃癌 磁共振成像 直方图分析 

分 类 号：R735.2[医药卫生—肿瘤] R445.2[医药卫生—临床医学]