Linking functional and structural brain images with multivariate network analyses: A novel application of the partial least square method

Kewei Chen, Eric M. Reiman, Zhongdan Huan, Richard John Caselli, Daniel Bandy, Napatkamon Ayutyanont, Gene E. Alexander

Research output: Contribution to journalArticle

43 Citations (Scopus)

Abstract

In this article, we introduce a multimodal multivariate network analysis to characterize the linkage between the patterns of information from the same individual's complementary brain images, and illustrate its potential by showing its ability to distinguish older from younger adults with greater power than several previously established methods. Our proposed method uses measurements from every brain voxel in each person's complementary co-registered images and uses the partial least square (PLS) algorithm to form a combined latent variable that maximizes the covariance among all of the combined variables. It represents a new way to calculate the singular value decomposition from the high-dimensional covariance matrix in a computationally feasible way. Analyzing fluorodeoxyglucose positron emission tomography (PET) and volumetric magnetic resonance imaging (MRI) images, this method distinguished 14 older adults from 15 younger adults (p = 4e- 12) with no overlap between groups, no need to correct for multiple comparisons, and greater power than the univariate Statistical Parametric Mapping (SPM), multimodal SPM or multivariate PLS analysis of either imaging modality alone. This technique has the potential to link patterns of information among any number of complementary images from an individual, to use other kinds of complementary complex datasets besides brain images, and to characterize individual state- or trait-dependent brain patterns in a more powerful way.

Original languageEnglish (US)
Pages (from-to)602-610
Number of pages9
JournalNeuroImage
Volume47
Issue number2
DOIs
StatePublished - Aug 15 2009

Fingerprint

Least-Squares Analysis
Multivariate Analysis
Brain
Young Adult
Positron-Emission Tomography
Magnetic Resonance Imaging

ASJC Scopus subject areas

  • Cognitive Neuroscience
  • Neurology

Cite this

Linking functional and structural brain images with multivariate network analyses : A novel application of the partial least square method. / Chen, Kewei; Reiman, Eric M.; Huan, Zhongdan; Caselli, Richard John; Bandy, Daniel; Ayutyanont, Napatkamon; Alexander, Gene E.

In: NeuroImage, Vol. 47, No. 2, 15.08.2009, p. 602-610.

Research output: Contribution to journalArticle

Chen, Kewei ; Reiman, Eric M. ; Huan, Zhongdan ; Caselli, Richard John ; Bandy, Daniel ; Ayutyanont, Napatkamon ; Alexander, Gene E. / Linking functional and structural brain images with multivariate network analyses : A novel application of the partial least square method. In: NeuroImage. 2009 ; Vol. 47, No. 2. pp. 602-610.
@article{1de160595b7a4fe982aec2f78e4da151,
title = "Linking functional and structural brain images with multivariate network analyses: A novel application of the partial least square method",
abstract = "In this article, we introduce a multimodal multivariate network analysis to characterize the linkage between the patterns of information from the same individual's complementary brain images, and illustrate its potential by showing its ability to distinguish older from younger adults with greater power than several previously established methods. Our proposed method uses measurements from every brain voxel in each person's complementary co-registered images and uses the partial least square (PLS) algorithm to form a combined latent variable that maximizes the covariance among all of the combined variables. It represents a new way to calculate the singular value decomposition from the high-dimensional covariance matrix in a computationally feasible way. Analyzing fluorodeoxyglucose positron emission tomography (PET) and volumetric magnetic resonance imaging (MRI) images, this method distinguished 14 older adults from 15 younger adults (p = 4e- 12) with no overlap between groups, no need to correct for multiple comparisons, and greater power than the univariate Statistical Parametric Mapping (SPM), multimodal SPM or multivariate PLS analysis of either imaging modality alone. This technique has the potential to link patterns of information among any number of complementary images from an individual, to use other kinds of complementary complex datasets besides brain images, and to characterize individual state- or trait-dependent brain patterns in a more powerful way.",
author = "Kewei Chen and Reiman, {Eric M.} and Zhongdan Huan and Caselli, {Richard John} and Daniel Bandy and Napatkamon Ayutyanont and Alexander, {Gene E.}",
year = "2009",
month = "8",
day = "15",
doi = "10.1016/j.neuroimage.2009.04.053",
language = "English (US)",
volume = "47",
pages = "602--610",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press Inc.",
number = "2",

}

TY - JOUR

T1 - Linking functional and structural brain images with multivariate network analyses

T2 - A novel application of the partial least square method

AU - Chen, Kewei

AU - Reiman, Eric M.

AU - Huan, Zhongdan

AU - Caselli, Richard John

AU - Bandy, Daniel

AU - Ayutyanont, Napatkamon

AU - Alexander, Gene E.

PY - 2009/8/15

Y1 - 2009/8/15

N2 - In this article, we introduce a multimodal multivariate network analysis to characterize the linkage between the patterns of information from the same individual's complementary brain images, and illustrate its potential by showing its ability to distinguish older from younger adults with greater power than several previously established methods. Our proposed method uses measurements from every brain voxel in each person's complementary co-registered images and uses the partial least square (PLS) algorithm to form a combined latent variable that maximizes the covariance among all of the combined variables. It represents a new way to calculate the singular value decomposition from the high-dimensional covariance matrix in a computationally feasible way. Analyzing fluorodeoxyglucose positron emission tomography (PET) and volumetric magnetic resonance imaging (MRI) images, this method distinguished 14 older adults from 15 younger adults (p = 4e- 12) with no overlap between groups, no need to correct for multiple comparisons, and greater power than the univariate Statistical Parametric Mapping (SPM), multimodal SPM or multivariate PLS analysis of either imaging modality alone. This technique has the potential to link patterns of information among any number of complementary images from an individual, to use other kinds of complementary complex datasets besides brain images, and to characterize individual state- or trait-dependent brain patterns in a more powerful way.

AB - In this article, we introduce a multimodal multivariate network analysis to characterize the linkage between the patterns of information from the same individual's complementary brain images, and illustrate its potential by showing its ability to distinguish older from younger adults with greater power than several previously established methods. Our proposed method uses measurements from every brain voxel in each person's complementary co-registered images and uses the partial least square (PLS) algorithm to form a combined latent variable that maximizes the covariance among all of the combined variables. It represents a new way to calculate the singular value decomposition from the high-dimensional covariance matrix in a computationally feasible way. Analyzing fluorodeoxyglucose positron emission tomography (PET) and volumetric magnetic resonance imaging (MRI) images, this method distinguished 14 older adults from 15 younger adults (p = 4e- 12) with no overlap between groups, no need to correct for multiple comparisons, and greater power than the univariate Statistical Parametric Mapping (SPM), multimodal SPM or multivariate PLS analysis of either imaging modality alone. This technique has the potential to link patterns of information among any number of complementary images from an individual, to use other kinds of complementary complex datasets besides brain images, and to characterize individual state- or trait-dependent brain patterns in a more powerful way.

UR - http://www.scopus.com/inward/record.url?scp=67349174191&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67349174191&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2009.04.053

DO - 10.1016/j.neuroimage.2009.04.053

M3 - Article

C2 - 19393744

AN - SCOPUS:67349174191

VL - 47

SP - 602

EP - 610

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

IS - 2

ER -