北京生物医学工程

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

Comparative study of voxel-basedand deformation-based morphometry in the identification of mild cognitive impairment

作者：周震景斌

单位：首都医科大学生物医学工程学院（北京 100069） 通信作者：周震。E-mail: zhouzhenbme@126.com

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

分类号：R318.04

出版年·卷·期（页码）：2021·40·5（494-498）

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(DBM)在检测轻度认知功能障碍(deformation-based morphometry ,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. Conclusion 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.

参考文献：

[1] Zhang W, Song LH, Yin XT, et al. Grey Matter Abnormalities in untreated hyperthyroidism: a voxel-based morphometry study using the DARTEL approach [J]. European Journal of Radiology, 2014, 83(1): 43-48.

[2] Yoon B, Shim Y.S, Hong Y.J, et al. Comparison of diffusion tensor imaging and voxel-based morphometry to detect white matter damage in alzheimers’s disease [J]. Journal of the Neurological Sciences, 2011, 302(1-2): 89-95. ?

[3] Chetelat G, Landeau B, Eustache F, et al. Using voxel-based morphometry to map the structural changes associated with rapid conversion in MCI:a longitudinal MRI study[J]. Neuroimage, 2005, 27(4): 934-946.

[4] Can D D, Richard T, Kuhl P K. Early grey-matter and white matter concentration in infancy predict later language skills: a whole brain voxel-based morphometry study [J]. Brain and Language, 2013, 124(1): 34-44. ?

[5] Gaser?C,?Nenadic I, Buchsbaum B R，et al. Deformation-based morphometry and its relation to conventional volumetry of brain lateral ventricles in MRI [J]. NeuroImage, 2001, 13(6): 1140-1145.

[6] Savio A, Grana M. Deformation based feature selection for computer aided diagnosis of alzheimer’s disease[J]. Expert Systems with Applications, 2013, 40(5): 1619-1628.

[7] Long ZQ,?Huang JC, Li B,?et al. A comparative atlas-based recognition of mild cognitive impairment with voxel-based morphometry?[J]. Frontiers in Neuroscience, 2018, 12：916.

[8] Long ZQ, Jing B, Guo R, et al. A brainnetome atlas based mild cognitive impairment identification using hurst exponent [J]. Frontiers in Aging Neuroscience, 2018, 10:103.

[9] Liu, H, Jing B, Han WJ et al. A?comparative texture analysis based on NECT and CECT?images to differentiate lung adenocarcinoma from squamous cell carcinoma [J]. Journal of Medical Systems,2019,.43(3):59.

[10] Ferreira?LK,?Diniz BS, Forlenza OV,?et al. Neurostructural predictors of Alzheimer's disease: A meta-analysis of VBM studies?[J].?Neurobiology of Aging,2011, 32(10):1733-1741.

[11] Nickl-Jockschat T, Kleiman A, Schulz JB, et al. Neuroanatomic changes and their association with cognitive decline in mild cognitive impairment: a meta-analysis [J]. Brain Structure & Function, 2012,217(1):115-125.

[12] Tapiola T,?Pennanen C, Tapiola M,?et al. MRI of hippocampus and entorhinal cortex in mild cognitive impairment: A follow-up study?[J]. Neurobiology of Aging,?2008, 29.1:31-38.

[13] Hirjak D,?Wolf RC, Remmele B,?et al. Hippocampal formation alterations differently contribute to autobiographic memory deficits in mild cognitive impairment and Alzheimer's disease?[J]. Hippocampus,2017, 27(6):702-715.

[14] Ren P,?Anthony M, Chapman BP,?et al. Amygdala functional connectivity is associated with locus of control in the context of cognitive aging?[J]. Neuropsychologia,?2017,99:199-206.

[15] Yao HX,?Liu Y, Zhou B,?et al. Decreased functional connectivity of the amygdala in Alzheimer's disease revealed by resting-state fMRI?[J]. European Journal of Radiology, 2013,82（9）:1531-1538.

[16] Fan Y,?Batmanghelich N,Clark CM,?et al. Spatial patterns of brain atrophy in MCI patients, identified via high-dimensional pattern classification, predict subsequent cognitive decline?[J]. Neuroimage,?2008,39（4）:1731-1743.

服务与反馈：

【文章下载】【加入收藏】

提示：您还未登录，请登录！点此登录