Learning brain connectivity of Alzheimer's disease by sparse inverse covariance estimation

Shuai Huang, Jing Li, Liang Sun, Jieping Ye, Adam Fleisher, Teresa Wu, Kewei Chen, Eric Reiman

Research output: Contribution to journalArticle

161 Citations (Scopus)

Abstract

Rapid advances in neuroimaging techniques provide great potentials for study of Alzheimer's disease (AD). Existing findings have shown that AD is closely related to alteration in the functional brain network, i.e., the functional connectivity between different brain regions. In this paper, we propose a method based on sparse inverse covariance estimation (SICE) to identify functional brain connectivity networks from PET data. Our method is able to identify both the connectivity network structure and strength for a large number of brain regions with small sample sizes. We apply the proposed method to the PET data of AD, mild cognitive impairment (MCI), and normal control (NC) subjects. Compared with NC, AD shows decrease in the amount of inter-region functional connectivity within the temporal lobe especially between the area around hippocampus and other regions and increase in the amount of connectivity within the frontal lobe as well as between the parietal and occipital lobes. Also, AD shows weaker between-lobe connectivity than within-lobe connectivity and weaker between-hemisphere connectivity, compared with NC. In addition to being a method for knowledge discovery about AD, the proposed SICE method can also be used for classifying new subjects, which makes it a suitable approach for novel connectivity-based AD biomarker identification. Our experiments show that the best sensitivity and specificity our method can achieve in AD vs. NC classification are 88% and 88%, respectively.

Original languageEnglish (US)
Pages (from-to)935-949
Number of pages15
JournalNeuroImage
Volume50
Issue number3
DOIs
StatePublished - Apr 15 2010

Fingerprint

Alzheimer Disease
Learning
Brain
Occipital Lobe
Parietal Lobe
Frontal Lobe
Temporal Lobe
Neuroimaging
Sample Size
Hippocampus
Biomarkers
Sensitivity and Specificity

Keywords

  • Alzheimer's
  • Biomarker
  • Brain connectivity
  • PET
  • Sparse inverse covariance

ASJC Scopus subject areas

  • Cognitive Neuroscience
  • Neurology

Cite this

Learning brain connectivity of Alzheimer's disease by sparse inverse covariance estimation. / Huang, Shuai; Li, Jing; Sun, Liang; Ye, Jieping; Fleisher, Adam; Wu, Teresa; Chen, Kewei; Reiman, Eric.

In: NeuroImage, Vol. 50, No. 3, 15.04.2010, p. 935-949.

Research output: Contribution to journalArticle

Huang, Shuai ; Li, Jing ; Sun, Liang ; Ye, Jieping ; Fleisher, Adam ; Wu, Teresa ; Chen, Kewei ; Reiman, Eric. / Learning brain connectivity of Alzheimer's disease by sparse inverse covariance estimation. In: NeuroImage. 2010 ; Vol. 50, No. 3. pp. 935-949.
@article{01117b5326924bb0822f9429480d99e2,
title = "Learning brain connectivity of Alzheimer's disease by sparse inverse covariance estimation",
abstract = "Rapid advances in neuroimaging techniques provide great potentials for study of Alzheimer's disease (AD). Existing findings have shown that AD is closely related to alteration in the functional brain network, i.e., the functional connectivity between different brain regions. In this paper, we propose a method based on sparse inverse covariance estimation (SICE) to identify functional brain connectivity networks from PET data. Our method is able to identify both the connectivity network structure and strength for a large number of brain regions with small sample sizes. We apply the proposed method to the PET data of AD, mild cognitive impairment (MCI), and normal control (NC) subjects. Compared with NC, AD shows decrease in the amount of inter-region functional connectivity within the temporal lobe especially between the area around hippocampus and other regions and increase in the amount of connectivity within the frontal lobe as well as between the parietal and occipital lobes. Also, AD shows weaker between-lobe connectivity than within-lobe connectivity and weaker between-hemisphere connectivity, compared with NC. In addition to being a method for knowledge discovery about AD, the proposed SICE method can also be used for classifying new subjects, which makes it a suitable approach for novel connectivity-based AD biomarker identification. Our experiments show that the best sensitivity and specificity our method can achieve in AD vs. NC classification are 88{\%} and 88{\%}, respectively.",
keywords = "Alzheimer's, Biomarker, Brain connectivity, PET, Sparse inverse covariance",
author = "Shuai Huang and Jing Li and Liang Sun and Jieping Ye and Adam Fleisher and Teresa Wu and Kewei Chen and Eric Reiman",
year = "2010",
month = "4",
day = "15",
doi = "10.1016/j.neuroimage.2009.12.120",
language = "English (US)",
volume = "50",
pages = "935--949",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press Inc.",
number = "3",

}

TY - JOUR

T1 - Learning brain connectivity of Alzheimer's disease by sparse inverse covariance estimation

AU - Huang, Shuai

AU - Li, Jing

AU - Sun, Liang

AU - Ye, Jieping

AU - Fleisher, Adam

AU - Wu, Teresa

AU - Chen, Kewei

AU - Reiman, Eric

PY - 2010/4/15

Y1 - 2010/4/15

N2 - Rapid advances in neuroimaging techniques provide great potentials for study of Alzheimer's disease (AD). Existing findings have shown that AD is closely related to alteration in the functional brain network, i.e., the functional connectivity between different brain regions. In this paper, we propose a method based on sparse inverse covariance estimation (SICE) to identify functional brain connectivity networks from PET data. Our method is able to identify both the connectivity network structure and strength for a large number of brain regions with small sample sizes. We apply the proposed method to the PET data of AD, mild cognitive impairment (MCI), and normal control (NC) subjects. Compared with NC, AD shows decrease in the amount of inter-region functional connectivity within the temporal lobe especially between the area around hippocampus and other regions and increase in the amount of connectivity within the frontal lobe as well as between the parietal and occipital lobes. Also, AD shows weaker between-lobe connectivity than within-lobe connectivity and weaker between-hemisphere connectivity, compared with NC. In addition to being a method for knowledge discovery about AD, the proposed SICE method can also be used for classifying new subjects, which makes it a suitable approach for novel connectivity-based AD biomarker identification. Our experiments show that the best sensitivity and specificity our method can achieve in AD vs. NC classification are 88% and 88%, respectively.

AB - Rapid advances in neuroimaging techniques provide great potentials for study of Alzheimer's disease (AD). Existing findings have shown that AD is closely related to alteration in the functional brain network, i.e., the functional connectivity between different brain regions. In this paper, we propose a method based on sparse inverse covariance estimation (SICE) to identify functional brain connectivity networks from PET data. Our method is able to identify both the connectivity network structure and strength for a large number of brain regions with small sample sizes. We apply the proposed method to the PET data of AD, mild cognitive impairment (MCI), and normal control (NC) subjects. Compared with NC, AD shows decrease in the amount of inter-region functional connectivity within the temporal lobe especially between the area around hippocampus and other regions and increase in the amount of connectivity within the frontal lobe as well as between the parietal and occipital lobes. Also, AD shows weaker between-lobe connectivity than within-lobe connectivity and weaker between-hemisphere connectivity, compared with NC. In addition to being a method for knowledge discovery about AD, the proposed SICE method can also be used for classifying new subjects, which makes it a suitable approach for novel connectivity-based AD biomarker identification. Our experiments show that the best sensitivity and specificity our method can achieve in AD vs. NC classification are 88% and 88%, respectively.

KW - Alzheimer's

KW - Biomarker

KW - Brain connectivity

KW - PET

KW - Sparse inverse covariance

UR - http://www.scopus.com/inward/record.url?scp=77249118678&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77249118678&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2009.12.120

DO - 10.1016/j.neuroimage.2009.12.120

M3 - Article

C2 - 20079441

AN - SCOPUS:77249118678

VL - 50

SP - 935

EP - 949

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

IS - 3

ER -