表观扩散系数灰度直方图在乳腺良恶性病变鉴别中的价值
Value of multi-parameter apparent diffusion coefficient histogram analysis in differential diagnosis of benign and malignant breast lesions

目的:探讨磁共振表观扩散系数(ADC)灰度直方图参数在乳腺良恶性病变鉴别中的价值。方法:对82例患者行乳腺常规、动态增强及扩散加权成像序列,后处理工作站直接获得ADC图。采用MaZda软件对病变ADC图像信息进行提取、分析,获得病变的灰度平均值、方差、偏度、峰度、第1百分位数、第10百分位数、第50百分位数、第90百分位数、第99百分位数,比较各参数在乳腺良恶性病变中的差异,采用ROC曲线分析各参数鉴别乳腺良恶性病变的效能。结果:乳腺恶性病变的灰度平均值、第10百分位数、第50百分位数、第90百分位数、第99百分位数均低于良性病变(P均<0.05),而恶性病变的偏度高于良性病变(P=0.001)。灰度平均值和第50百分位数具有较高的鉴别诊断价值,其AUC分别为0.816、0.829,相应的诊断阈值分别为133.34、128.00,敏感度和特异度分别为0.738、0.820和0.786、0.760。结论:ADC灰度直方图参数在乳腺良恶性病变的鉴别中具有较大价值。
Aim: To investigate the differential diagnostic value of apparent diffusion coefficient(ADC)map histogram analysis in benign and malignant breast lesions.Methods:A total of 82 patients with breast tumors underwent T1WI,T2WI,dynamic contrast-enhanced T1WI and DWI.The ADC map was obtained from post-processing station. The histogram metrics, including mean, variance, skewness, kurtosis, as well as the 1st,10th,50th,90th and 99th percentiles were measured by the MaZda software. All parameters were compared between malignant and benign breast lesions. Receiver operating characteristic curve was used to analyze and compare the ability of these parameters in differentiation of benign and malignant breast lesions.Results: The mean, the 10th,50th,90th,and 99th percentiles were significantly lower in malignant tumors than those in benign tumors(P<0.05), and the skewness was significantly higher in malignant tumors than that in benign tumors(P=0.001). The mean and the 50th percentile had higher area under the receiver operating characteristic curve(0.816 and 0.829),and their optimal cutoff values were 133.34 and 128.00,with sensibility and specificity of 0.738, 0.820 and 0.786, 0.760.Conclusion: The parameters of ADC histogram analysis could be helpful for differentiation between benign and malignant breast tumors