基于AlexNet模型的AD分类

R445.2; 阿尔兹海默症(Alzheimer’s disease,AD)一经发现难以逆转,早期诊断对延缓AD的病程发展非常重要. 虽然深度卷积网络近年来在图像识别领域有着大量突出的表现,但将从自然图像中训练得到的二维经典的深度网络直接运用到三维的结构磁共振影像(structural magnetic resonance imaging,sMRI)上进行AD疾病状态的分类还存在一些问题. 基于194 例 AD、123 例晚期轻度认知障碍(late mild cognitive impairment,LMCI)与105 例正常老化(normal control,NC)的sMRI,运用特征迁移学...

Full description

Saved in:
Bibliographic Details
Published in北京工业大学学报 Vol. 46; no. 1; pp. 68 - 74
Main Authors 张柏雯, 林岚, 吴水才
Format Journal Article
LanguageChinese
Published 北京工业大学生命科学与生物工程学院,北京,100124 2020
Subjects
结构磁共振图像
特征提取
structural magnetic resonance imaging ( sMRI)
feature extract
三维特征重组
阿尔兹海默症(AD)
3D features reconstruct
AlexNet
迁移学习
transform learning
Alzheimer’s disease ( AD )
Online AccessGet full text
ISSN0254-0037
DOI10.11936/bjutxb2018070029

Cover

Abstract R445.2; 阿尔兹海默症(Alzheimer’s disease,AD)一经发现难以逆转,早期诊断对延缓AD的病程发展非常重要. 虽然深度卷积网络近年来在图像识别领域有着大量突出的表现,但将从自然图像中训练得到的二维经典的深度网络直接运用到三维的结构磁共振影像(structural magnetic resonance imaging,sMRI)上进行AD疾病状态的分类还存在一些问题. 基于194 例 AD、123 例晚期轻度认知障碍(late mild cognitive impairment,LMCI)与105 例正常老化(normal control,NC)的sMRI,运用特征迁移学习的方法,从经典的深度卷积模型——AlexNet中提取各阶段受试者的图像特征,并对所提特征进行三维重组,再运用最大池化、主成分分析等方式降维,并运用向前序列选择方法对各分类组进行特征选择,最后运用支持向量机建立分类模型,实现AD、LMCI与NC的分类. 在AlexNet的三、四、五层卷积层,AD与NC的分类准确率分别为89. 93%、91. 28%、87. 25%,AD与 LMCI的分类结果分别为80. 77%、76. 92%、78. 21%,NC与LMCI的分类结果分别为72. 46%、75. 45%、73. 65%. 结果证明,通过经典卷积网络获得的特征,经过三维重组,能够较好地对AD实现分类.
AbstractList R445.2; 阿尔兹海默症(Alzheimer’s disease,AD)一经发现难以逆转,早期诊断对延缓AD的病程发展非常重要. 虽然深度卷积网络近年来在图像识别领域有着大量突出的表现,但将从自然图像中训练得到的二维经典的深度网络直接运用到三维的结构磁共振影像(structural magnetic resonance imaging,sMRI)上进行AD疾病状态的分类还存在一些问题. 基于194 例 AD、123 例晚期轻度认知障碍(late mild cognitive impairment,LMCI)与105 例正常老化(normal control,NC)的sMRI,运用特征迁移学习的方法,从经典的深度卷积模型——AlexNet中提取各阶段受试者的图像特征,并对所提特征进行三维重组,再运用最大池化、主成分分析等方式降维,并运用向前序列选择方法对各分类组进行特征选择,最后运用支持向量机建立分类模型,实现AD、LMCI与NC的分类. 在AlexNet的三、四、五层卷积层,AD与NC的分类准确率分别为89. 93%、91. 28%、87. 25%,AD与 LMCI的分类结果分别为80. 77%、76. 92%、78. 21%,NC与LMCI的分类结果分别为72. 46%、75. 45%、73. 65%. 结果证明,通过经典卷积网络获得的特征,经过三维重组,能够较好地对AD实现分类.
Abstract_FL Alzheimer’s disease ( AD) generally results in irreversible brain damages. Early diagnosis of disease plays an important role in preventing the progression of AD. Deep convolutional neural networks (CNN) have achieved prominent performance in the field of natural image recognition, while some problems exist in applying a classic CNN model on 3D MRI for AD classification. To address these issues, with 194 AD subjects, late mild cognitive impairment ( LMCI) 123 subjects and 105 normal control(NC) subjects, a hybrid computational strategy was proposed based on a pre-trained AlexNet CNN model and sMRI for AD classification. The feature presentation of pre-trained network was efficiently transferred to AD classification task by using transfer learning, 3D features reconstruction, feature reduction using Max pooling and principal component analysis ( PCA ) , and selection feature using sequential forward search and ( SFS) method. Then, support vector machines ( SVM) was applied to classification. The classification accuracy values on conv3, conv4, conv5 layers of AlexNet were 89. 93%,91. 28%, and 87. 25% of AD/NC, respectively, 80. 77%,76. 92%, and 78. 21% of AD/LMCI, respectively, 72. 46%,75. 45%, and 73. 65% of NC/LMCI, respectively. Results show that features extracted from a classic CNN model and their 3D reconstruction can achieve good performance on AD classification.
Author 吴水才
张柏雯
林岚
AuthorAffiliation 北京工业大学生命科学与生物工程学院,北京,100124
AuthorAffiliation_xml – name: 北京工业大学生命科学与生物工程学院,北京,100124
Author_FL LIN Lan
ZHANG Baiwen
WU Shuicai
Author_FL_xml – sequence: 1
  fullname: ZHANG Baiwen
– sequence: 2
  fullname: LIN Lan
– sequence: 3
  fullname: WU Shuicai
Author_xml – sequence: 1
  fullname: 张柏雯
– sequence: 2
  fullname: 林岚
– sequence: 3
  fullname: 吴水才
BookMark eNotj7tKA0EUQKeIYEzyATb2Fqv3zt2dR7nEJwTTaB32zs4ElzABN8G1FxEEP8BCsbO1jb_jmt9Q0ep053C2RCfOoxdiG2EP0ZLa52q5aFgCGtAA0nZEF2SWJgCkN8Wgri8ZIJVWI1FX7LQvq8_VYz7zzZlffL29ts8P66fb_KC9v1u_f_TFRihmtR_8sycujg7PhyfJaHx8OsxHSY2Q2kR7Bg6ki8DeFuCcVawtG8lKhSwwQQkWM6eMKo0hwkxJ5VInXYnM0lBP7P55r4sYijidVPPlVfwpTria3pTN75AEQABL36smSeU
ClassificationCodes R445.2
ContentType Journal Article
Copyright Copyright © Wanfang Data Co. Ltd. All Rights Reserved.
Copyright_xml – notice: Copyright © Wanfang Data Co. Ltd. All Rights Reserved.
DBID 2B.
4A8
92I
93N
PSX
TCJ
DOI 10.11936/bjutxb2018070029
DatabaseName Wanfang Data Journals - Hong Kong
WANFANG Data Centre
Wanfang Data Journals
万方数据期刊 - 香港版
China Online Journals (COJ)
China Online Journals (COJ)
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
DocumentTitle_FL Efficient Alzheimer’s Disease Classification Based on AlexNet Model
EndPage 74
ExternalDocumentID bjgydxxb202001009
GrantInformation_xml – fundername: 北京市自然科学基金-海淀原始创新联合基金资助项目; 北京市教委计划一般项目; 国家科技支撑计划课题
  funderid: (L182010); (KM201810005033); (2015BAI02B03)
GroupedDBID -03
2B.
4A8
5XA
5XD
92H
92I
93N
ABJNI
ACGFS
ADMLS
ALMA_UNASSIGNED_HOLDINGS
CCEZO
CEKLB
CW9
P2P
PSX
TCJ
TGT
U1G
U5M
ID FETCH-LOGICAL-s1049-7eb0bf37afbe9a0cc96b79b82b66f5fb30d0915c686d883315626c4c2cd1bb283
ISSN 0254-0037
IngestDate Thu May 29 03:59:35 EDT 2025
IsPeerReviewed false
IsScholarly true
Issue 1
Keywords 结构磁共振图像
特征提取
structural magnetic resonance imaging ( sMRI)
feature extract
三维特征重组
阿尔兹海默症(AD)
3D features reconstruct
AlexNet
迁移学习
transform learning
Alzheimer’s disease ( AD )
Language Chinese
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-s1049-7eb0bf37afbe9a0cc96b79b82b66f5fb30d0915c686d883315626c4c2cd1bb283
PageCount 7
ParticipantIDs wanfang_journals_bjgydxxb202001009
PublicationCentury 2000
PublicationDate 2020
PublicationDateYYYYMMDD 2020-01-01
PublicationDate_xml – year: 2020
  text: 2020
PublicationDecade 2020
PublicationTitle 北京工业大学学报
PublicationTitle_FL Journal of Beijing University of Technology
PublicationYear 2020
Publisher 北京工业大学生命科学与生物工程学院,北京,100124
Publisher_xml – name: 北京工业大学生命科学与生物工程学院,北京,100124
SSID ssib004297133
ssib051370302
ssj0039890
ssib001129165
ssib002263171
Score 2.2153287
Snippet R445.2; 阿尔兹海默症(Alzheimer’s disease,AD)一经发现难以逆转,早期诊断对延缓AD的病程发展非常重要. 虽然深度卷积网络近年来在图像识别领域有着大量突出的表现,但将从自然图像中训练得到的二维经典的深度网络直接运用到三维的结构磁共振影像(structural magnetic...
SourceID wanfang
SourceType Aggregation Database
StartPage 68
Title 基于AlexNet模型的AD分类
URI https://d.wanfangdata.com.cn/periodical/bjgydxxb202001009
Volume 46
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1Lb9QwELZKe4ED4imeVYWYYyBO4tfR2WRVIdoLrdRbFWeToh4Wie5KpWeEkJD4ARxA3LhyLX-HpX-DmcTdtWgPpRfLmoztsSfOfHY8Y8aeyqquKjQ9kdOpjLJKpJE2UkRNpZUxtUtSQb7DG5tyfTt7sSN2lq7Ewaml6cQ9q4_O9Su5jFaRhnolL9n_0Oy8UiRgHvWLKWoY0wvpGEoBZgi5hTKjVJfkr7LZTKCUYDVY3nGUoHMoFRjkyGxBNK1BS6LlHPI8RKjd0wEY5eu0A6LkCqzoKJqqQYrNwKouU4CVQUaCxlLiVJNd6QHYmJ6gsHoIpQGTgx0uWCQJabrq8gQbCAubGPKMWPLTjDagi3C7Ionnr9ZlxcfRyUg8ahCHpCAKMhge8HTFdRkwK5LFmqAJReNOwz0vhZ1FngTf5XNkSwYUnar37_bfY1xLRxSvJzQefv80nCS9JegvC_KYor-I6Ky1Mt19Om5_Ojl0CYVSU_SXdGGa5wcm3f7eu9EhMdEZuM5jdSXBhRGaohVbbLx8tYDACOB48Oca4TVCRB7iD8UXEF_wlL7585BrqdF-X9J31p8EIFGf_yto5-g2bqvxXoDJtm6w634xtWb7mXGTLR29vsWuBSE2b7PV2bfj38ef_bz48-P77Ounky_vbTH7-OHk5687bHtYbg3WI38nSHTAcTEbqcbFrk1V1brGVHFdG-mUcTpxUraidWk8QgQsaqnliO7R5ojvZZ3VST3iziGWvsuWx2_GzT22JgQOB2LEVGuTtRnXziRaUzzB2rTciPvsie_drp_zB7tn9PDgIkwP2VXK9zt3j9jy5O20eYxYduJWvfr-ArWra30
linkProvider EBSCOhost
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=%E5%9F%BA%E4%BA%8EAlexNet%E6%A8%A1%E5%9E%8B%E7%9A%84AD%E5%88%86%E7%B1%BB&rft.jtitle=%E5%8C%97%E4%BA%AC%E5%B7%A5%E4%B8%9A%E5%A4%A7%E5%AD%A6%E5%AD%A6%E6%8A%A5&rft.au=%E5%BC%A0%E6%9F%8F%E9%9B%AF&rft.au=%E6%9E%97%E5%B2%9A&rft.au=%E5%90%B4%E6%B0%B4%E6%89%8D&rft.date=2020&rft.pub=%E5%8C%97%E4%BA%AC%E5%B7%A5%E4%B8%9A%E5%A4%A7%E5%AD%A6%E7%94%9F%E5%91%BD%E7%A7%91%E5%AD%A6%E4%B8%8E%E7%94%9F%E7%89%A9%E5%B7%A5%E7%A8%8B%E5%AD%A6%E9%99%A2%2C%E5%8C%97%E4%BA%AC%2C100124&rft.issn=0254-0037&rft.volume=46&rft.issue=1&rft.spage=68&rft.epage=74&rft_id=info:doi/10.11936%2Fbjutxb2018070029&rft.externalDocID=bjgydxxb202001009
thumbnail_s http://utb.summon.serialssolutions.com/2.0.0/image/custom?url=http%3A%2F%2Fwww.wanfangdata.com.cn%2Fimages%2FPeriodicalImages%2Fbjgydxxb%2Fbjgydxxb.jpg