T1WI增强直方图分析鉴别血管瘤型脑膜瘤和非典型脑膜瘤OA北大核心CSTPCD
Differential diagnosis of angiomatous meningioma and atypical meningioma based on contrast enhanced T1-weighted images histogram analysis
目的 本研究拟探讨结构MRI特征及T1WI增强图像直方图分析在鉴别非典型脑膜瘤(atypical meningioma,AtM)和血管瘤型脑膜瘤(angiomatous meningioma,AnM)中的价值.材料与方法 回顾性分析经组织病理证实的AtM(n=40)和AnM(n=30)的临床、影像和病理资料.使用MaZda软件在轴位T1WI增强图像上对肿瘤进行逐层勾画并获得肿瘤强化区的直方图参数.结构MRI特征采用卡方检验或者Fisher's精确检验对比,采用独立样本t检验或Mann-Whitney U检验对两组间的直方图参数进行比较,两组间的诊断效能由受试者工作特征(receiver operating characteristic,ROC)曲线来评估.结果 AtM组中 肿瘤坏死的发生率(75.0%)明显高于AnM组(36.7%)(P=0.001).AnM 组中T1WI增强图像的平均值(P=0.003)和第 1(P<0.001)、第10(P=0.001)、第50百分位数(P=0.009)均大于AtM,差异有统计学意义.ROC曲线分析显示,肿瘤坏死+融合直方图参数对二者的鉴别诊断效能最优,其曲线下面积、敏感度、特异度、准确率、阳性预测值和阴性预测值分别为0.858(0.753~0.930)、95.00%、66.67%、82.86%、79.20%和90.90%.结论 常规MRI特征和T1WI增强直方图分析有助于术前无创鉴别AnM和AtM,其中以肿瘤坏死+融合直方图参数的诊断效能最高.
Objective:To investigate the value of structural MRI features and enhanced T1-weighted images histogram analysis in the differential diagnosis of atypical meningioma(AtM)and angiomatous meningioma(AnM).Materials and Methods:The clinical,imaging and pathological data of AtM(n=40)and AnM(n=30)were collected retrospectively.The tumor was delineated layer-by-layer on axial enhanced T1-weighted images by MaZda software and histogram parameters of tumor enhancement areas were obtained.Structural MRI characteristics were compared using the chi-square test or Fisher's exact test.Histogram parameters were compared between the two groups using independent samples t-tests or Mann-Whitney U-tests,and diagnostic efficacy between the two groups was assessed by receiver operating characteristic(ROC)curves.Results:The incidence of tumor necrosis was significantly higher in the AtM group(75.0%)than in the AnM group(36.7%)(P=0.001).The mean(P=0.003),1st percentile(P<0.001),10th percentile(P=0.001),and 50th percentile(P=0.009)of the AnM were greater than those of the AtM.ROC curve analysis showed that tumor necrosis + combined histogram parameters had the optimal differential diagnostic efficacy between the two,with area under the curve,sensitivity,specificity,accuracy,positive predictive value,and negative predictive value of 0.858(0.753-0.930),95.00%,66.67%,82.86%,79.20%,and 90.90%.Conclusions:The conventional MRI features and histogram analysis based on enhanced T1-weighted images is helpful in the preoperative non-invasive differentiation of AnM and AtM,of which the diagnostic efficacy is highest in the tumor necrosis + combined histogram parameters.
韩涛;刘显旺;蒋健;周凤瑜;董文洁;张斌;周俊林
兰州大学第二医院放射科,兰州 730030||兰州大学第二临床医学院,兰州 730030||甘肃省医学影像重点实验室,兰州 730030||医学影像人工智能甘肃省国际科技合作基地,兰州 730030兰州大学第二医院放射科,兰州 730030||甘肃省医学影像重点实验室,兰州 730030||医学影像人工智能甘肃省国际科技合作基地,兰州 730030
临床医学
脑膜瘤血管瘤型脑膜瘤非典型脑膜瘤磁共振成像直方图分析
meningiomaangiomatous meningiomaatypical meningiomamagnetic resonance imaginghistogram analysis
《磁共振成像》 2024 (003)
37-42 / 6
National Natural Science Foundation of China(No.81772006,82071872);Science and Technology Planning Project of Gansu Province(No.21YF5FA123);China International Medical Exchange Foundation-SKY Imaging Research Fund(No.Z-2014-07-2101). 国家自然科学基金面上项目(编号:81772006、82071872);甘肃省科技计划项目(编号:21YF5FA123);中华国际医学交流基金会-SKY影像科研基金项目(编号:Z-2014-07-2101)
评论