DCE-MRI参数直方图术前评估胃癌分化程度的研究  

Preoperative evaluation of gastric cancer differentiation using DCE-MRI parameter histograms

作　　者：金建国 高丽渊 王象萍 潘孝本 JIN Jianguo;GAO Liyuan;WANG Xiangping;PAN Xiaoben(Department of Radiology,Linping District,the Second Affiliated Hospital of Zhejiang University School of Medicine,Hangzhou,Zhejiang 311100,China)

机构地区：[1]浙江大学医学院附属第二医院临平院区放射科,浙江杭州311100 [2]杭州师范大学医学院,浙江杭州311100

出　　处：《中华全科医学》2023年第10期1728-1731,共4页Chinese Journal of General Practice

基　　金：国家自然科学基金常规面上项目(82070610);2021年杭州市医药卫生科技项目(B20220841)。

摘　　要：目的本研究以术前胃癌患者为研究对象,旨在分析DCE-MRI直方图定量参数对术前评估胃癌分化程度的诊断价值。方法选取2020年5月—2023年5月浙江大学医学院附属第二医院临平院区收治的128例胃癌患者为研究对象。术前行DCE-MRI检查,获得DCE-MRI定量参数直方图,包括血管外细胞外间隙容积比(V_(e))、容积转运常数(K_(trans))、速率常数(K_(ep))的偏度、熵、峰度、平均值及10%、90%(P_(10)、P_(90))。分析上述参数对胃癌分化程度的评估价值。结果低分化组V_(e)的P_(10)和K_(trans)及K_(ep)的熵、平均值、P_(10)、P_(90)高于中分化组和高分化组(均P<0.05);低分化组K_(ep)的偏度、峰度低于中分化组和高分化组(均P<0.05)。Logistic回归分析结果显示K_(trans)的熵、平均值、P_(10)、P_(90)和K_(ep)的偏度、熵、平均值、P_(10)、P_(90)水平异常升高是影响胃癌分化程度的危险因素(均P<0.05)。ROC结果显示,K_(trans)的熵、平均值、P_(10)、P_(90)和K_(ep)的偏度、熵、平均值、P_(10)、P_(90)诊断低分化胃癌的AUC分别为0.739、0.806、0.812、0.758、0.714、0.726、0.786、0.761、0.749。结论DCE-MRI定量参数直方图对术前胃癌分化程度具有较高的诊断预测价值,其中K_(trans)的P_(10)诊断效能最佳。Objective To analyze the diagnostic value of quantitative parameters of DCE-MRI histogram in preoperative evaluating the degree of differentiation of gastric cancer.Methods A total of 128 patients with gastric cancer admitted to the Second Affiliated Hospital of Zhejiang University School of Medicine from May 2020 to May 2023 were selected as the study objects.DCE-MRI was performed before surgery to obtain the histogram of quantitative parameters of DCE-MRI,including the skewness,entropy,kurtosis,average value and 10%,90%(P_(10),P_(90))of extravascular extracellular space volume ratio(V_(e)),volume transport constant(K_(trans))and rate constant(K_(ep)).The value of the above parameters in evaluating the differentiation degree of gastric cancer was analyzed.Results The entropy,mean value,P_(10)and P_(90)of P_(10),K_(trans)and K_(ep)of V_(e)in low differentiation group were higher than those in middle differentiation group and high differentiation group(all P<0.05).The skewness and kurtosis of K_(ep)in low differentiation group were lower than those in middle differentiation group and high differentiation group(all P<0.05).Logistic regression analysis showed that abnormal increases in K_(trans)entropy,mean value,skewness,entropy,mean value,P_(10),P_(90)and K_(ep)levels were risk factors affecting the differentiation degree of gastric cancer(all P<0.05).ROC results showed that the AUC of K_(trans)entropy,mean value,P_(10),P_(90)and K_(ep)skewness,entropy,mean value,P_(10)and P_(90)in the diagnosis of poorly differentiated gastric cancer were 0.739,0.806,0.812,0.758,0.714,0.726,0.786,0.761 and 0.749,respectively.Conclusion DCE-MRI quantitative parameter histogram has high diagnostic value in predicting the differentiation degree of gastric cancer,and K_(trans)P_(10)has the best diagnostic efficacy.

关 键 词：磁共振动态对比增强成像 直方图分析 胃癌 分化程度 

分 类 号：R735.2[医药卫生—肿瘤] R730.4[医药卫生—临床医学]