基于ADC图的全域直方图分析在鉴别交界性和恶性上皮性卵巢肿瘤的价值探究

Value of Global Histogram Analysis Based on ADC Map in Differentiating Borderline and Malignant Epithelial Ovarian Tumors

目的探讨全域表观扩散系数(ADC)直方图定量参数鉴别交界性上皮性卵巢肿瘤(BEOT)和恶性上皮性卵巢肿瘤(MEOT)的价值。方法回顾性分析经病理证实的67例MEOT和34例BEOT的患者临床影像资料。采用FireVoxel软件在肿瘤轴位ADC图上逐层勾画感兴趣区(ROI)并进行全域直方图分析,获得8个直方图参数。使用Mann-Whitney U检验对两组间ADC直方图参数进行比较。采用受试者工作特征曲线(ROC)曲线下面积(AUC)来评估ADC直方图参数的诊断效能。结果两组间定量参数比较,第1、10、50、90、99百分位数、偏度、熵及手动勾画ADC值有统计学差异(P<0.001),而峰度无统计学意义。两组间第10百分位数的AUC为0.851,敏感度、特异度、准确率分别为85.07%,88.24%,86.13%。结论基于ADC图的全域直方图分析在BEOT和MEOT的鉴别中具有临床价值,值得推广。

Objective To explore the value of the histogram quantitative parameter of the global apparent diffusion coefficient(ADC)histogram in identifying borderline epithelial ovarian tumor(BEOT)from malignant epithelial ovarian tumor(MEOT).Methods The clinical imaging data of 67 patients with MEOT and 34 patients with BEOT confirmed by pathology were analyzed retrospectively.Firevoxel software was used to draw the region of interest(ROI)layer by layer on the axial ADC image of the tumor,and the global histogram analysis was carried out to obtain 8 histogram parameters.The ADC histogram parameters between two groups was tested by Mann Whitney U test.The area under the receiver operating characteristic(ROC)curve was used to evaluate the diagnostic efficiency of ADC histogram parameters.Results Compared with the quantitative parameters between the two groups,the 1st,10th,50th,90th,99th percentile,skewness,entropy and manual sketching ADC values had statistical differences(P<0.001),but kurtosis had no statistical significance.The area under the curve of the 10th percentile between the two groups was 0.851,and the sensitivity,specificity and accuracy were 85.07%,88.24%and 86.13%respectively.Conclusion The global histogram analysis based on ADC map in the differentiation of borderline and malignant epithelial ovarian tumors has clinical value,hence this method is worthwhile to use in clinic practice.