VTSRM:一种基于SVM-RFE和MRMR的AD MRI医学图像分类方法
|
摘要
为了准确地识别阿尔兹海默症(Alzheimer's Disease,AD),轻度认知障碍(Mild Cognitive Impairment,MCI)和正常个体(Normal Controls,NC),论文实现了一种基于SVM-RFE和MRMR的AD MRI医学图像分类方法 VTSRM。该方法首先提取出MRI医学图像的纹理特征和形态学特征,然后利用基于支持向量机递归特征消除算法(SVM-RFE)和最小冗余最大相关(MRMR)技术的特征选择算法SRM选择出最优特征子集,并使用SVM分类算法对AD,MCI,NC进行分类。美国公共阿尔茨海默病神经影像学数据集上的实验证明了论文方法的有效性。
In order to accurately distinguish Alzheimer's disease(AD),Mild Cognitive Impairment(MCI)and normal controls(NC),this paper implements a method VTSRM based on SVM-RFE and MRMR for AD MRI medical image classification.The method first extracts the texture features and morphological features of MRI medical images,and then selects the optimal feature subset by using the feature selection algorithm SRM which based on Support Vector Machine Recursive Feature Elimination Algorithm(SVM-RFE)and Minimum Redundant Maximum Correlation(MRMR). Last,SVM classification algorithm is used to classify AD,MCI,NC. Experiments on the neuroimaging dataset of public Alzheimer's disease in the United States demonstrate the effectiveness of the proposed method.
In order to accurately distinguish Alzheimer's disease(AD),Mild Cognitive Impairment(MCI)and normal controls(NC),this paper implements a method VTSRM based on SVM-RFE and MRMR for AD MRI medical image classification.The method first extracts the texture features and morphological features of MRI medical images,and then selects the optimal feature subset by using the feature selection algorithm SRM which based on Support Vector Machine Recursive Feature Elimination Algorithm(SVM-RFE)and Minimum Redundant Maximum Correlation(MRMR). Last,SVM classification algorithm is used to classify AD,MCI,NC. Experiments on the neuroimaging dataset of public Alzheimer's disease in the United States demonstrate the effectiveness of the proposed method.
引文
[1]李润辉.阿尔兹海默病的研究现状[J].沈阳医学院学报,2013,15(3):129-133.LI Runhui. Research status of Alzheimer's disease[J].Journal of Shenyang Medical College,2013,15(3):129-133.
[2]何毅,王华丽,苏敏莹,等.轻度认知损害和阿尔茨海默病的大脑灰质体积及其与记忆功能的关系[J].中国神经精神疾病杂志,2012,38(7):411-416.HE Yi,WANG Huali,SU Minying,et al. Relationship between the brain gray matter volume and memory performance in mild cognitive impairment and Alzheimer’s disease:a voxel-based morphometric study[J]. Chinese Journal of Nervous and Mental Diseases,2012,38(7):411-416.
[3] Alzheimer's Association. 2015 Alzheimer's disease facts and figures.[J]. Alzheimers&Dementia,2015,11(3):332-384.
[4]Petersen R C,Smith G E,Waring S C,et al. Mild cognitive impairment:clinical characterization and outcome.[J]. Archives of Neurology,1999,56(3):303-308.
[5]Busatto G F,Diniz B S,Zanetti M V. Voxel-based morphometry in Alzheimer's disease[J]. Expert review of neurotherapeutics,2008,8(11):1691-1702.
[6]Beheshti I,Demirel H,Matsuda H,et al. Classification of Alzheimer's disease and prediction of mild cognitive impairment-to-Alzheimer's conversion from structural magnetic resource imaging using feature ranking and a genetic algorithm[J]. Computers in biology and medicine,2017,83(4):109-119.
[7]Zhe X,Yi D,Tian L,et al. Brain MR Image Classification for Alzheimer's Disease Diagnosis Based on Multifeature Fusion[J]. Computational and Mathematical Methods in Medicine,2017,(5):1-13.
[8]李晓.基于脑MRI特征提取和分类的应用与研究[D].成都:西南交通大学,2016.LI Xiao. The application and research of feature extraction and classification method based on MRI[D]. Chengdu:Southwest Jiaotong University,2016.
[9]Guyon I,Weston J,Barnhill S,et al. Gene selection for cancer classification using support vector machines[J].Machine learning,2002,46(1-3):389-422.
[10]Peng H,Long F,Ding C. Feature selection based on mutual information criteria of max-dependency,max-relevance,and min-redundancy[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(8):1226-1238.
[11]Jun L,Lu X U,Long L. Feature Selection Method Based on SVM-RFE and Particle Swarm Optimization[J]. Journal of Chinese Computer Systems,2015,36(8):1865-1868.
[12]张健,曾卫明,杨嘉君.基于SVM-RFE的静息态fMRI分类研究及应用[J].现代计算机(专业版),2017,(20):5-9.ZHANG Jian,ZENG Weiming,YANG Jiajun. Research on Classification of Rest fMRI Data Based on SVM-RFE[J]. Modern Computer,2017,(20):5-9.
[13]Lin X,Li C,Zhang Y,et al. Selecting feature subsets based on svm-rfe and the overlapping ratio with applications in bioinformatics[J]. Molecules,2017,23(1):52-61.
[14]Mao Y,Yang Y. A feature selection method based on effective range and SVM-RFE[J]. International Journal of Wireless and Mobile Computing,2018,15(2):105-112.
[15]Ashburner J. A fast diffeomorphic image registration algorithm[J]. Neuroimage,2007,38(1):95-113.
[16]Haralick R M,Shanmugam K. Textural features for image classification[J]. IEEE Transactions on systems,man,and cybernetics,1973(6):610-621.
[17]洪继光.灰度-梯度共生矩阵纹理分析方法[J].自动化学报,1984,10(01):22-25.HONG Jiguang. Gray Level-gradient cooccurrence matrix texture analysis method[J]. Acta Automatica Sinica,1984,10(01):22-25.
[2]何毅,王华丽,苏敏莹,等.轻度认知损害和阿尔茨海默病的大脑灰质体积及其与记忆功能的关系[J].中国神经精神疾病杂志,2012,38(7):411-416.HE Yi,WANG Huali,SU Minying,et al. Relationship between the brain gray matter volume and memory performance in mild cognitive impairment and Alzheimer’s disease:a voxel-based morphometric study[J]. Chinese Journal of Nervous and Mental Diseases,2012,38(7):411-416.
[3] Alzheimer's Association. 2015 Alzheimer's disease facts and figures.[J]. Alzheimers&Dementia,2015,11(3):332-384.
[4]Petersen R C,Smith G E,Waring S C,et al. Mild cognitive impairment:clinical characterization and outcome.[J]. Archives of Neurology,1999,56(3):303-308.
[5]Busatto G F,Diniz B S,Zanetti M V. Voxel-based morphometry in Alzheimer's disease[J]. Expert review of neurotherapeutics,2008,8(11):1691-1702.
[6]Beheshti I,Demirel H,Matsuda H,et al. Classification of Alzheimer's disease and prediction of mild cognitive impairment-to-Alzheimer's conversion from structural magnetic resource imaging using feature ranking and a genetic algorithm[J]. Computers in biology and medicine,2017,83(4):109-119.
[7]Zhe X,Yi D,Tian L,et al. Brain MR Image Classification for Alzheimer's Disease Diagnosis Based on Multifeature Fusion[J]. Computational and Mathematical Methods in Medicine,2017,(5):1-13.
[8]李晓.基于脑MRI特征提取和分类的应用与研究[D].成都:西南交通大学,2016.LI Xiao. The application and research of feature extraction and classification method based on MRI[D]. Chengdu:Southwest Jiaotong University,2016.
[9]Guyon I,Weston J,Barnhill S,et al. Gene selection for cancer classification using support vector machines[J].Machine learning,2002,46(1-3):389-422.
[10]Peng H,Long F,Ding C. Feature selection based on mutual information criteria of max-dependency,max-relevance,and min-redundancy[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(8):1226-1238.
[11]Jun L,Lu X U,Long L. Feature Selection Method Based on SVM-RFE and Particle Swarm Optimization[J]. Journal of Chinese Computer Systems,2015,36(8):1865-1868.
[12]张健,曾卫明,杨嘉君.基于SVM-RFE的静息态fMRI分类研究及应用[J].现代计算机(专业版),2017,(20):5-9.ZHANG Jian,ZENG Weiming,YANG Jiajun. Research on Classification of Rest fMRI Data Based on SVM-RFE[J]. Modern Computer,2017,(20):5-9.
[13]Lin X,Li C,Zhang Y,et al. Selecting feature subsets based on svm-rfe and the overlapping ratio with applications in bioinformatics[J]. Molecules,2017,23(1):52-61.
[14]Mao Y,Yang Y. A feature selection method based on effective range and SVM-RFE[J]. International Journal of Wireless and Mobile Computing,2018,15(2):105-112.
[15]Ashburner J. A fast diffeomorphic image registration algorithm[J]. Neuroimage,2007,38(1):95-113.
[16]Haralick R M,Shanmugam K. Textural features for image classification[J]. IEEE Transactions on systems,man,and cybernetics,1973(6):610-621.
[17]洪继光.灰度-梯度共生矩阵纹理分析方法[J].自动化学报,1984,10(01):22-25.HONG Jiguang. Gray Level-gradient cooccurrence matrix texture analysis method[J]. Acta Automatica Sinica,1984,10(01):22-25.