Principal component analysis in mild and moderate Alzheimer's disease - A novel approach to clinical diagnosis

Marco Pagani, Dario Salmaso, Guido Rodriguez, Davide Nardo, Flavio Nobili

Research output: Contribution to journalArticle

Abstract

Principal component analysis (PCA) provides a method to explore functional brain connectivity. The aim of this study was to identify regional cerebral blood flow (rCBF) distribution differences between Alzheimer's disease (AD) patients and controls (CTR) by means of volume of interest (VOI) analysis and PCA. Thirty-seven CTR, 30 mild AD (mildAD) and 27 moderate AD (modAD) subjects were investigated using single photon emission computed tomography with 99mTc-hexamethylpropylene amine oxime. Analysis of covariance (ANCOVA), PCA, and discriminant analysis (DA) were performed on 54 VOIs. VOI analysis identified in both mildAD and modAD subjects a decreased rCBF in six regions. PCA in mildAD subjects identified four principal components (PCs) in which the correlated VOIs showed a decreased level of rCBF, including regions that are typically affected early in the disease. In five PCs, including parietal-temporal-limbic cortex, and hippocampus, a significantly lower rCBF in correlated VOIs was found in modAD subjects. DA significantly discriminated the groups. The percentage of subjects correctly classified was 95, 70, and 81 for CTR, mildAD and modAD groups, respectively. PCA highlighted, in mildAD and modAD, relationships not evident when brain regions are considered as independent of each other, and it was effective in discriminating groups. These findings may allow neurophysiological inferences to be drawn regarding brain functional connectivity in AD that might not be possible with univariate analysis.

Original languageEnglish
Pages (from-to)8-14
Number of pages7
JournalPsychiatry Research - Neuroimaging
Volume173
Issue number1
DOIs
Publication statusPublished - Jul 15 2009

Fingerprint

Cerebrovascular Circulation
Principal Component Analysis
Regional Blood Flow
Alzheimer Disease
Discriminant Analysis
Brain
Oximes
Temporal Lobe
Single-Photon Emission-Computed Tomography
Amines
Hippocampus

Keywords

  • Computerised Brain Atlas
  • Dementia
  • Discriminant analysis
  • SPECT
  • Volume of interest analysis

ASJC Scopus subject areas

  • Psychiatry and Mental health
  • Radiology Nuclear Medicine and imaging
  • Neuroscience (miscellaneous)

Cite this

Principal component analysis in mild and moderate Alzheimer's disease - A novel approach to clinical diagnosis. / Pagani, Marco; Salmaso, Dario; Rodriguez, Guido; Nardo, Davide; Nobili, Flavio.

In: Psychiatry Research - Neuroimaging, Vol. 173, No. 1, 15.07.2009, p. 8-14.

Research output: Contribution to journalArticle

Pagani, Marco ; Salmaso, Dario ; Rodriguez, Guido ; Nardo, Davide ; Nobili, Flavio. / Principal component analysis in mild and moderate Alzheimer's disease - A novel approach to clinical diagnosis. In: Psychiatry Research - Neuroimaging. 2009 ; Vol. 173, No. 1. pp. 8-14.
@article{20bd66074eaa440389a336a1fec1ed29,
title = "Principal component analysis in mild and moderate Alzheimer's disease - A novel approach to clinical diagnosis",
abstract = "Principal component analysis (PCA) provides a method to explore functional brain connectivity. The aim of this study was to identify regional cerebral blood flow (rCBF) distribution differences between Alzheimer's disease (AD) patients and controls (CTR) by means of volume of interest (VOI) analysis and PCA. Thirty-seven CTR, 30 mild AD (mildAD) and 27 moderate AD (modAD) subjects were investigated using single photon emission computed tomography with 99mTc-hexamethylpropylene amine oxime. Analysis of covariance (ANCOVA), PCA, and discriminant analysis (DA) were performed on 54 VOIs. VOI analysis identified in both mildAD and modAD subjects a decreased rCBF in six regions. PCA in mildAD subjects identified four principal components (PCs) in which the correlated VOIs showed a decreased level of rCBF, including regions that are typically affected early in the disease. In five PCs, including parietal-temporal-limbic cortex, and hippocampus, a significantly lower rCBF in correlated VOIs was found in modAD subjects. DA significantly discriminated the groups. The percentage of subjects correctly classified was 95, 70, and 81 for CTR, mildAD and modAD groups, respectively. PCA highlighted, in mildAD and modAD, relationships not evident when brain regions are considered as independent of each other, and it was effective in discriminating groups. These findings may allow neurophysiological inferences to be drawn regarding brain functional connectivity in AD that might not be possible with univariate analysis.",
keywords = "Computerised Brain Atlas, Dementia, Discriminant analysis, SPECT, Volume of interest analysis",
author = "Marco Pagani and Dario Salmaso and Guido Rodriguez and Davide Nardo and Flavio Nobili",
year = "2009",
month = "7",
day = "15",
doi = "10.1016/j.pscychresns.2008.07.016",
language = "English",
volume = "173",
pages = "8--14",
journal = "Psychiatry Research - Neuroimaging",
issn = "0925-4927",
publisher = "Elsevier Ireland Ltd",
number = "1",

}

TY - JOUR

T1 - Principal component analysis in mild and moderate Alzheimer's disease - A novel approach to clinical diagnosis

AU - Pagani, Marco

AU - Salmaso, Dario

AU - Rodriguez, Guido

AU - Nardo, Davide

AU - Nobili, Flavio

PY - 2009/7/15

Y1 - 2009/7/15

N2 - Principal component analysis (PCA) provides a method to explore functional brain connectivity. The aim of this study was to identify regional cerebral blood flow (rCBF) distribution differences between Alzheimer's disease (AD) patients and controls (CTR) by means of volume of interest (VOI) analysis and PCA. Thirty-seven CTR, 30 mild AD (mildAD) and 27 moderate AD (modAD) subjects were investigated using single photon emission computed tomography with 99mTc-hexamethylpropylene amine oxime. Analysis of covariance (ANCOVA), PCA, and discriminant analysis (DA) were performed on 54 VOIs. VOI analysis identified in both mildAD and modAD subjects a decreased rCBF in six regions. PCA in mildAD subjects identified four principal components (PCs) in which the correlated VOIs showed a decreased level of rCBF, including regions that are typically affected early in the disease. In five PCs, including parietal-temporal-limbic cortex, and hippocampus, a significantly lower rCBF in correlated VOIs was found in modAD subjects. DA significantly discriminated the groups. The percentage of subjects correctly classified was 95, 70, and 81 for CTR, mildAD and modAD groups, respectively. PCA highlighted, in mildAD and modAD, relationships not evident when brain regions are considered as independent of each other, and it was effective in discriminating groups. These findings may allow neurophysiological inferences to be drawn regarding brain functional connectivity in AD that might not be possible with univariate analysis.

AB - Principal component analysis (PCA) provides a method to explore functional brain connectivity. The aim of this study was to identify regional cerebral blood flow (rCBF) distribution differences between Alzheimer's disease (AD) patients and controls (CTR) by means of volume of interest (VOI) analysis and PCA. Thirty-seven CTR, 30 mild AD (mildAD) and 27 moderate AD (modAD) subjects were investigated using single photon emission computed tomography with 99mTc-hexamethylpropylene amine oxime. Analysis of covariance (ANCOVA), PCA, and discriminant analysis (DA) were performed on 54 VOIs. VOI analysis identified in both mildAD and modAD subjects a decreased rCBF in six regions. PCA in mildAD subjects identified four principal components (PCs) in which the correlated VOIs showed a decreased level of rCBF, including regions that are typically affected early in the disease. In five PCs, including parietal-temporal-limbic cortex, and hippocampus, a significantly lower rCBF in correlated VOIs was found in modAD subjects. DA significantly discriminated the groups. The percentage of subjects correctly classified was 95, 70, and 81 for CTR, mildAD and modAD groups, respectively. PCA highlighted, in mildAD and modAD, relationships not evident when brain regions are considered as independent of each other, and it was effective in discriminating groups. These findings may allow neurophysiological inferences to be drawn regarding brain functional connectivity in AD that might not be possible with univariate analysis.

KW - Computerised Brain Atlas

KW - Dementia

KW - Discriminant analysis

KW - SPECT

KW - Volume of interest analysis

UR - http://www.scopus.com/inward/record.url?scp=67349272711&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67349272711&partnerID=8YFLogxK

U2 - 10.1016/j.pscychresns.2008.07.016

DO - 10.1016/j.pscychresns.2008.07.016

M3 - Article

C2 - 19443186

AN - SCOPUS:67349272711

VL - 173

SP - 8

EP - 14

JO - Psychiatry Research - Neuroimaging

JF - Psychiatry Research - Neuroimaging

SN - 0925-4927

IS - 1

ER -