基于灰度共生矩阵的MRI纹理分析在椎管内脊膜瘤和神经鞘瘤鉴别诊断中的应用价值  

Application value of MRI texture analysis based on GLCM in differential diagnosis of intraspinal meningioma and schwannoma

作　　者：梁洁 杜昕 王现亮[1] 蒲如剑 朱万平 LIANG Jie;DU Xin;WANG Xianliang;PU Rujian;ZHU Wanping(Department of Radiology,Weifang People's Hospital,Weifang 261041,China;School of Medical Imaging,Weifang Medical University,Weifang 261053,China;Department of Spinal Surgery,Weifang People's Hospital,Weifang 261041,China)

机构地区：[1]潍坊市人民医院放射科,潍坊261041 [2]潍坊医学院医学影像学院,潍坊261053 [3]潍坊市人民医院脊柱外科,潍坊261041

出　　处：《磁共振成像》2022年第8期84-87,共4页Chinese Journal of Magnetic Resonance Imaging

摘　　要：目的探讨基于灰度共生矩阵(gray-level co-occurrence matrix,GLCM)的MR图像纹理分析在鉴别椎管内脊膜瘤和神经鞘瘤中的临床应用价值。材料与方法回顾性分析经病理证实的椎管内神经鞘瘤患者病例32例、脊膜瘤患者病例26例,利用Image J软件在T2WI、对比增强T1WI矢状位图像中选取肿瘤最大层面勾画感兴趣区(region of interest,ROI),提取病灶的GLCM纹理参数,比较两组肿瘤各参数的差异,并评价各参数的诊断效能。结果T2WI序列中能量、对比、相关、逆差矩和熵在两组间差异均有统计学意义(P<0.05),神经鞘瘤组的能量、相关、逆差矩小于脊膜瘤组,对比和熵大于脊膜瘤组;对比增强T1WI序列中能量、对比、相关和熵在两组间差异有统计学意义(P<0.05),神经鞘瘤组的能量、相关小于脊膜瘤组,对比和熵大于脊膜瘤组,逆差矩组间差异无统计学意义(P>0.05)。受试者工作特征(receiver operating characteristic,ROC)曲线分析显示,T2WI序列中的熵及对比增强T1WI序列中的能量诊断效能最佳。采用logistic回归分析纹理参数联合诊断效能发现较单一参数均有所提高。结论基于GLCM的MRI图像纹理分析在椎管内脊膜瘤和神经鞘瘤两者的鉴别诊断中有一定的临床价值。Objective: To investigate the clinical value of MRI texture analysis based on gray-level co-occurrence matrix(GLCM) in differentiating intraspinal meningioma and schwannoma. Materials and Methods: Thirty-two cases of intraspinal schwannoma and26 cases of meningioma confirmed by pathology were analyzed retrospectively. The region of interest(ROI) of the largest layer of the tumor was selected in T2WI and contrast-enhanced T1WI sagittal images by using imageJ software, and the GLCM texture parameters of the lesions were extracted. Results: The differences of tumor parameters between the two groups were compared, and the diagnostic efficiency of each parameter was evaluated. There was a significant difference between the two groups(P<0.05);there was significant difference in energy, contrast, correlation and entropy between the two groups in contrast-enhanced T1WI sequence(P<0.05). The energy and correlation of schwannoma group were less than that of meningioma group, the contrast and entropy were greater than that of meningioma group, and there was no significant difference between inverse gap groups(P>0.05). ROC curve analysis showed that the entropy in T2WI sequence and the energy diagnosis efficiency in contrast-enhanced T1WI sequence were the best. The joint diagnosis of texture parameters by logistic regression analysis has improved the diagnosis efficiency compared with that of single parameter. Conclusions: MRI texture analysis based on GLCM has certain clinical value in the differential diagnosis of intraspinal meningioma and schwannoma.

关 键 词：磁共振成像 纹理分析 脊膜瘤 神经鞘瘤 灰度共生矩阵 

分 类 号：R445.2[医药卫生—影像医学与核医学] R739.42[医药卫生—诊断学] R730.264[医药卫生—临床医学]