Alzheimer's disease diagnosis in individual subjects using structural MR images

Validation studies

Research output: Contribution to journalArticle

300 Citations (Scopus)

Abstract

Objective: To develop and validate a tool for Alzheimer's disease (AD) diagnosis in individual subjects using support vector machine (SVM)-based classification of structural MR (sMR) images. Background: Libraries of sMR scans of clinically well characterized subjects can be harnessed for the purpose of diagnosing new incoming subjects. Methods: One hundred ninety patients with probable AD were age- and gender-matched with 190 cognitively normal (CN) subjects. Three different classification models were implemented: Model I uses tissue densities obtained from sMR scans to give STructural Abnormality iNDex (STAND)-score; and Models II and III use tissue densities as well as covariates (demographics and Apolipoprotein E genotype) to give adjusted-STAND (aSTAND)-score. Data from 140 AD and 140 CN were used for training. The SVM parameter optimization and training were done by four-fold cross validation (CV). The remaining independent sample of 50 AD and 50 CN was used to obtain a minimally biased estimate of the generalization error of the algorithm. Results: The CV accuracy of Model II and Model III aSTAND-scores was 88.5% and 89.3%, respectively, and the developed models generalized well on the independent test data sets. Anatomic patterns best differentiating the groups were consistent with the known distribution of neurofibrillary AD pathology. Conclusions: This paper presents preliminary evidence that application of SVM-based classification of an individual sMR scan relative to a library of scans can provide useful information in individual subjects for diagnosis of AD. Including demographic and genetic information in the classification algorithm slightly improves diagnostic accuracy.

Original languageEnglish (US)
Pages (from-to)1186-1197
Number of pages12
JournalNeuroImage
Volume39
Issue number3
DOIs
StatePublished - Feb 1 2008

Fingerprint

Validation Studies
Alzheimer Disease
Libraries
Demography
Apolipoproteins E
Genotype
Pathology
Support Vector Machine

Keywords

  • Alzheimer's
  • Classification
  • Diagnosis
  • Support vector machines

ASJC Scopus subject areas

  • Cognitive Neuroscience
  • Neurology

Cite this

@article{eb295620a5d749cf8ab0bcf666e62ff8,
title = "Alzheimer's disease diagnosis in individual subjects using structural MR images: Validation studies",
abstract = "Objective: To develop and validate a tool for Alzheimer's disease (AD) diagnosis in individual subjects using support vector machine (SVM)-based classification of structural MR (sMR) images. Background: Libraries of sMR scans of clinically well characterized subjects can be harnessed for the purpose of diagnosing new incoming subjects. Methods: One hundred ninety patients with probable AD were age- and gender-matched with 190 cognitively normal (CN) subjects. Three different classification models were implemented: Model I uses tissue densities obtained from sMR scans to give STructural Abnormality iNDex (STAND)-score; and Models II and III use tissue densities as well as covariates (demographics and Apolipoprotein E genotype) to give adjusted-STAND (aSTAND)-score. Data from 140 AD and 140 CN were used for training. The SVM parameter optimization and training were done by four-fold cross validation (CV). The remaining independent sample of 50 AD and 50 CN was used to obtain a minimally biased estimate of the generalization error of the algorithm. Results: The CV accuracy of Model II and Model III aSTAND-scores was 88.5{\%} and 89.3{\%}, respectively, and the developed models generalized well on the independent test data sets. Anatomic patterns best differentiating the groups were consistent with the known distribution of neurofibrillary AD pathology. Conclusions: This paper presents preliminary evidence that application of SVM-based classification of an individual sMR scan relative to a library of scans can provide useful information in individual subjects for diagnosis of AD. Including demographic and genetic information in the classification algorithm slightly improves diagnostic accuracy.",
keywords = "Alzheimer's, Classification, Diagnosis, Support vector machines",
author = "Vemuri, {Prashanthi D} and Gunter, {Jeffrey L.} and Senjem, {Matthew L.} and Whitwell, {Jennifer Lynn} and Kantarci, {Kejal M} and Knopman, {David S} and Boeve, {Bradley F} and Petersen, {Ronald Carl} and Jack, {Clifford R Jr.}",
year = "2008",
month = "2",
day = "1",
doi = "10.1016/j.neuroimage.2007.09.073",
language = "English (US)",
volume = "39",
pages = "1186--1197",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press Inc.",
number = "3",

}

TY - JOUR

T1 - Alzheimer's disease diagnosis in individual subjects using structural MR images

T2 - Validation studies

AU - Vemuri, Prashanthi D

AU - Gunter, Jeffrey L.

AU - Senjem, Matthew L.

AU - Whitwell, Jennifer Lynn

AU - Kantarci, Kejal M

AU - Knopman, David S

AU - Boeve, Bradley F

AU - Petersen, Ronald Carl

AU - Jack, Clifford R Jr.

PY - 2008/2/1

Y1 - 2008/2/1

N2 - Objective: To develop and validate a tool for Alzheimer's disease (AD) diagnosis in individual subjects using support vector machine (SVM)-based classification of structural MR (sMR) images. Background: Libraries of sMR scans of clinically well characterized subjects can be harnessed for the purpose of diagnosing new incoming subjects. Methods: One hundred ninety patients with probable AD were age- and gender-matched with 190 cognitively normal (CN) subjects. Three different classification models were implemented: Model I uses tissue densities obtained from sMR scans to give STructural Abnormality iNDex (STAND)-score; and Models II and III use tissue densities as well as covariates (demographics and Apolipoprotein E genotype) to give adjusted-STAND (aSTAND)-score. Data from 140 AD and 140 CN were used for training. The SVM parameter optimization and training were done by four-fold cross validation (CV). The remaining independent sample of 50 AD and 50 CN was used to obtain a minimally biased estimate of the generalization error of the algorithm. Results: The CV accuracy of Model II and Model III aSTAND-scores was 88.5% and 89.3%, respectively, and the developed models generalized well on the independent test data sets. Anatomic patterns best differentiating the groups were consistent with the known distribution of neurofibrillary AD pathology. Conclusions: This paper presents preliminary evidence that application of SVM-based classification of an individual sMR scan relative to a library of scans can provide useful information in individual subjects for diagnosis of AD. Including demographic and genetic information in the classification algorithm slightly improves diagnostic accuracy.

AB - Objective: To develop and validate a tool for Alzheimer's disease (AD) diagnosis in individual subjects using support vector machine (SVM)-based classification of structural MR (sMR) images. Background: Libraries of sMR scans of clinically well characterized subjects can be harnessed for the purpose of diagnosing new incoming subjects. Methods: One hundred ninety patients with probable AD were age- and gender-matched with 190 cognitively normal (CN) subjects. Three different classification models were implemented: Model I uses tissue densities obtained from sMR scans to give STructural Abnormality iNDex (STAND)-score; and Models II and III use tissue densities as well as covariates (demographics and Apolipoprotein E genotype) to give adjusted-STAND (aSTAND)-score. Data from 140 AD and 140 CN were used for training. The SVM parameter optimization and training were done by four-fold cross validation (CV). The remaining independent sample of 50 AD and 50 CN was used to obtain a minimally biased estimate of the generalization error of the algorithm. Results: The CV accuracy of Model II and Model III aSTAND-scores was 88.5% and 89.3%, respectively, and the developed models generalized well on the independent test data sets. Anatomic patterns best differentiating the groups were consistent with the known distribution of neurofibrillary AD pathology. Conclusions: This paper presents preliminary evidence that application of SVM-based classification of an individual sMR scan relative to a library of scans can provide useful information in individual subjects for diagnosis of AD. Including demographic and genetic information in the classification algorithm slightly improves diagnostic accuracy.

KW - Alzheimer's

KW - Classification

KW - Diagnosis

KW - Support vector machines

UR - http://www.scopus.com/inward/record.url?scp=37849014569&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=37849014569&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2007.09.073

DO - 10.1016/j.neuroimage.2007.09.073

M3 - Article

VL - 39

SP - 1186

EP - 1197

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

IS - 3

ER -