期刊文献+

磁共振后扣带回纹理特征分析在老年健康人群、轻度认知障碍与阿尔茨海默病患者鉴别诊断中的价值 被引量:5

Differential diagnostic value of texture analysis of posterior cingulate region on T1WI among the healthy,mild cognitive impairment and Alzheimer disease in older adults
暂未订购
导出
摘要 目的比较健康老年人(healthy elderly,HE)、轻度认识障碍(mild cognitive impairment,MCI)老年患者、阿尔茨海默病(alzheimer disease,AD)老年患者在后扣带回磁共振成像T1WI序列图像纹理特征差异,为临床诊断提供参考。方法选取2015年1月-2017年8月阿尔茨海默病神经影像学中心数据库168例健康老年人、233例轻度认识障碍老年患者及101例阿尔茨海默病老年患者T1WI图像进行后扣带回纹理特征分析,比较三组间角二阶矩、对比度、自相关、逆差距、熵5个纹理特征参数的差异,同时用多元有序logistic回归分析法建立诊断模型,绘制受试者工作特征曲线评价模型预测能力。结果HE与MCI、AD三组之间在角二阶矩(P=0.000)、对比度(P=0.001)、逆差距(P=0.000)和熵(P=0.000)这4个参数上的差异均有统计学意义,但自相关(P=0.095)不存在统计学差异;有序多分类logistic回归分析提示逆差距和简易智力状态检查量表评分可以作为鉴别三者的变量。基于logistic回归建立的诊断分析模型的受试者工作特征曲线下面积分别为0.894(HE vsMCI-AD)、0.995(AD vs HE-MCI)。结论基于磁共振成像T1WI序列图像的后扣带回纹理特征分析在HE、MCI与AD患者鉴别中有一定价值。 Objective To compare the difference of the texture features of the posterior cingulate on T1WI among healthy elderly(HE),the elderly patients with mild cognitive impairment(MCI)and the elderly patients with Alzheimer disease(AD).Methods Theimaging data were obtained from Alzheimer’s disease neuroimaging initiative(ADNI)database from January 2015 to August 2017.The texture analysis were performed based on T1WI among 168 HEs,233 patients with MCI and 101 patients with AD,and the texture parameters included Angular Second Moment(ASM),Contrast,Correlation,Inverse Difference Moment(IDM),and Entropy.The difference in texture parameter was compared between the three groups,and regression model was established by oridinal logistic regression analysis.Receiver operation characteristic(ROC)curve was used to evaluate the diagnostic efficacy.Results There were significant differences in the texture ASM,Contrast,IDM and Entropy among HE,MCI and AD(PASM=0.000,Pcontrast=0.001,PIDM=0.000 and Pentropy=0.000,respectively),while the texture Correlation presented no significant difference among the three goups(Pcorrelation=0.095).Oridinal Logistic regression analysis demonstrated that the texture IDM and Mini-mental State Examination score could be considered as the specific variables for the diagnosis of AD from HE and MCI.The area under the curve for equations established based on logistic regression was 0.894(HE vs MCI-AD)and 0.995(AD vs HE-MCI).Conclusion The texture analysis of the posterior cingulate on T1WI could primarily be used to differentiate HE,MCI and AD.
作者 王波涛 招云春 陈志晔 WANG Botao;ZHAO Yunchun;CHEN Zhiye(Department of Radiology,Hainan Hospital of Chinese PLA General Hospital,Sanya 572013,Hainan Province,China;Cadre Ward,Chinese PLA 920 Hospital,Kunming 650032,Yunnan Province,China;Department of Radiology,the First Medical Center,Chinese PLA General Hospital,Beijing 100853,China)
出处 《解放军医学院学报》 CAS 2020年第7期675-678,682,共5页 Academic Journal of Chinese PLA Medical School
关键词 轻度认识障碍 阿尔茨海默病 纹理分析 磁共振成像 鉴别诊断 mild cognitive impairment Alzheimer disease texture analysis magnetic resonance imaging differential diagnosis
  • 相关文献

参考文献11

二级参考文献97

  • 1于同刚,戴嘉中,冯晓源.原发性中枢神经系统淋巴瘤的MRI及^1H-MRS特点[J].临床放射学杂志,2005,24(8):668-672. 被引量:65
  • 2刘杰,林一帆,张沥,胡家露,曹晓燕,樊代明.图象自动分析检测MG7抗原表达预测胃癌高危价值探讨[J].中华预防医学杂志,1996,30(5):286-288. 被引量:13
  • 3陈卫中,潘晓平,倪宗瓒.logistic回归模型在ROC分析中的应用[J].中国卫生统计,2007,24(1):22-24. 被引量:59
  • 4宇传华 徐勇勇.ROC分析的基本原理[J].中华流行病学杂志,1998,19(2):413-415.
  • 5J A Modestino,J Zhang.A markov random field model based approach to image interpretation[J].IEEE Tran On Pattern Analysis and Machine Intelligence,1992,14(6):606-615.
  • 6N Kamath,K.Sunil Kumar,U B Desai.Joint segmentation and image interpretation using hidden Markov models[A].Proc of the Int Conf on Pattern Recognition[C].Brisbane,Australia,1998,2:1840-1842.
  • 7Belhadj Ziad,Bouhlel Nizar,Sevestre Ghalila Sylvie,Boussema Mohamed Rached.Heterogeneous SAR Texture Characterization By Means Of Markov Random Fields[A].IEEE 2000 International Geoscience and Remote Sensing Symposium Proceedings (IGARSS′2000)[C].Honolulu Hawaii,2000,2:579-581.
  • 8Rupert D Paget.Nonparametric Markov Random Field Models for Natural Texture Images[D].The University of Queensland,1999.
  • 9S C Liew,H Lim,L K Kwoh,G K Tay.Texture analysis of SAR images[A].IEEE 1995 International Geoscience and Remote Sensing Symposium Proceedings (IGARSS′1995)[C].Firenze,Italy,1995,2:1412-1414.
  • 10Robert M Haralick,K Shanmugam,Its′hak Dinstein.Texture features for image classification[J].IEEE Trans on Systems,Man and Cybernetics,1973,3(6):610-621.

共引文献363

同被引文献76

引证文献5

二级引证文献15

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部