期刊文献+

基于体素的和基于形变的形态学测量在轻度认知障碍识别上的比较研究 被引量:1

Comparative study of voxel-based and deformation-based morphometry in the identification of mild cognitive impairment
下载PDF
导出
摘要 目的比较基于体素的形态学测量(voxel-based morphometry,VBM)和基于形变的形态学测量(deformation-based morphometry,DBM)在检测轻度认知功能障碍(mild cognitive impairment,MCI)灰质异常及相应分类识别性能上的差异,为结构态分析方法的选择提供依据。方法利用VBM和DBM对27例MCI患者及30例健康对照的磁共振结构像进行分析,分别统计比较获得相应的组间结构异常脑区,并将异常脑区作为分类特征构建相应的MCI诊断识别模型,最终通过评价异常脑区的空间分布特征及分类识别准确率来评估两种方法的差异。结果VBM和DBM均发现MCI患者在海马、海马旁回、杏仁核、岛叶等脑区发生结构改变,但VBM方法还在额中回、颞中回等脑区发现异常。VBM确定的结构异常得到了86.0%的最佳准确度,而DBM方法的准确度为77.2%,虽然在性能表现上稍差,但发现的特征与VBM的最优特征具有一致性。结论VBM方法可以发现更多的MCI结构异常,而DBM方法则能发现具有较强敏感性的结构异常,因而提示在磁共振结构像研究中应将两者结合应用。 Objective To make comparisons between voxel-based morphometry(VBM)and deformation-based morphometry(DBM)in investigating gray matter abnormalities and corresponding identification performances for mild cognitive impairment(MCI),which may provide evidences for method selection in structural analysis studies.Methods The structural magnetic resonance imaging data of 27 MCI patients and 30 normal controls(NC)were analyzed with VBM and DBM methods,and the group differences were detected by statistical analysis,which was then used to construct the classification model for MCI.The spacial patterns of the detected abnormalities and the classification accuracies would be compared between two methods.Results Both VBM and DBM methods discovered structural abnormalities in regions such as parahippocampal gyrus,hippocampus,amygdala and insula.In addition,VBM also detected structural changes in regions like middle frontal gyrus and middle temporal gyrus.Furthermore,the proposed classification model for VBM achieved the best accuracy of 86.0%,while DBM model got an accuracy of 77.2%.Although DBM performed slightly worse than VBM,it detected consistent features as the optimal ones that VBM detected.Conclusions More structural abnormalities can be detected by VBM method in MCI patients,while DBM method can discover structural abnormalities with relatively high sensitivity,which suggests both methods should be combined together to study structural alterations.
作者 周震 景斌 ZHOU Zhen;JING Bin(School of Biomedical Engineering,Capital Medical University,Beijing 100069)
出处 《北京生物医学工程》 2021年第5期494-498,511,共6页 Beijing Biomedical Engineering
基金 北京市教委科技计划(KM202010025025)资助。
关键词 基于体素的形态学测量 基于形变的形态学测量 磁共振成像 轻度认知障碍 模式分类 voxel-based morphometry deformation-based morphometry magnetic resonance imaging mild cognitive impairment pattern classification
  • 相关文献

同被引文献6

引证文献1

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