摘要
目的:探讨MR T_2WI直方图分析在髓母细胞瘤复发风险评估中的应用价值。方法:将32例经手术病理证实的后颅窝髓母细胞瘤患者纳入研究,所有患者术后随访36个月,未见复发者纳入无复发组,发现复发者纳入复发组。在T_2WI图像中选择肿瘤最大层面,应用MaZda软件在病灶内勾画ROI进行灰度直方图分析,获得灰度平均值、方差、偏度、峰度及第1、第10、第50、第90和第99百分位数共9个参数的测量值并进行统计学分析,对两组间差异有统计学意义的参数,采用ROC曲线分析来评价其对髓母细胞瘤复发的诊断效能。结果:直方图分析提取的9个参数中,复发组的灰度方差小于无复发组,组间差异有统计学意义(P<0.05),两组间其它各项参数值的差异均无统计学意义(P>0.05)。灰度方差的ROC曲线下面积为0.725,鉴别肿瘤复发风险的临界值为376.26,诊断敏感度和特异度分别为60.0%和99.9%。结论:MR T_2WI直方图分析可作为评估髓母细胞瘤复发风险的重要辅助手段。
Objective:To explore the role of T 2-weighted MRI histogram in the risk assessment of medulloblastoma recurrence.Methods:Clinical and imaging data of 32 patients with pathologically confirmed medulloblastoma were retrospective analyzed.These patients were divided into two groups,which were defined as the recurrent group and the non-recurrent group.Using the software Mazda,region of interest were drawn in the tumors on T 2-weighted MR axial images on which the tumors diameter was maximum.Then the histogram parameters including average value,variance,skewness,kurtosis,and 1,10,50,90 and 99 percentile.The histogram parameters were analyzed statistically to find out which one has significant difference between the two groups,and then ROC curve analysis was used to assess its diagnostic efficiency in medulloblastoma recurrence.Results:As for the 9 parameters extracted from histogram,the differences of the variance between the two groups had statistical signi-ficance(P<0.05).Area under the ROC curve was 0.725.The sensitivity and specificity of variance were 60.0%and 99.9%.Conclusion:T 2-weighted MRI histogram analysis can be used as an important supplementary method to assess the risk of medulloblastoma recurrence.
作者
吕青青
张勇
程敬亮
朱晨迪
汪卫建
许珂
LV Qing-qing;ZHANG Yong;CHENG Jing-liang;ZHU Chendi;WANG Weijian;XU Ke(Department of MRI,the First Affiliated Hospital of Zhengzhou University,Zhengzhou 450052,China)
出处
《放射学实践》
北大核心
2018年第8期829-832,共4页
Radiologic Practice
基金
2016年河南省医学科技攻关项目(201602030)
关键词
髓母细胞瘤
脑肿瘤
肿瘤复发
直方图分析
磁共振成像
T2加权成像
Medulloblastoma
Brain neoplasms
Tumor recurrence
Histogram analysis
Magnetic resonance imaging
T 2-weighted imaging
