Resting-state multi-spectrum functional connectivity networks for identification of MCI patients

Chong Yaw Wee, Pew Thian Yap, Kevin Denny, Jeffrey N. Browndyke, Guy G. Potter, Kathleen A. Welsh-Bohmer, Lihong Wang, Dinggang Shen

Research output: Contribution to journalArticle

78 Citations (Scopus)

Abstract

In this paper, a high-dimensional pattern classification framework, based on functional associations between brain regions during resting-state, is proposed to accurately identify MCI individuals from subjects who experience normal aging. The proposed technique employs multi-spectrum networks to characterize the complex yet subtle blood oxygenation level dependent (BOLD) signal changes caused by pathological attacks. The utilization of multi-spectrum networks in identifying MCI individuals is motivated by the inherent frequency-specific properties of BOLD spectrum. It is believed that frequency specific information extracted from different spectra may delineate the complex yet subtle variations of BOLD signals more effectively. In the proposed technique, regional mean time series of each region-of-interest (ROI) is band-pass filtered (0:025 ≤ f ≤ 0:100 Hz) before it is decomposed into five frequency sub-bands. Five connectivity networks are constructed, one from each frequency sub-band. Clustering coefficient of each ROI in relation to the other ROIs are extracted as features for classification. Classification accuracy was evaluated via leave-one-out cross-validation to ensure generalization of performance. The classification accuracy obtained by this approach is 86.5%, which is an increase of at least 18.9% from the conventional full-spectrum methods. A cross-validation estimation of the generalization performance shows an area of 0.863 under the receiver operating characteristic (ROC) curve, indicating good diagnostic power. It was also found that, based on the selected features, portions of the prefrontal cortex, orbitofrontal cortex, temporal lobe, and parietal lobe regions provided the most discriminant information for classification, in line with results reported in previous studies. Analysis on individual frequency sub-bands demonstrated that different sub-bands contribute differently to classification, providing extra evidence regarding frequency-specific distribution of BOLD signals. Our MCI classification framework, which allows accurate early detection of functional brain abnormalities, makes an important positive contribution to the treatment management of potential AD patients.

Original languageEnglish
Article numbere37828
JournalPLoS One
Volume7
Issue number5
DOIs
Publication statusPublished - 2012 May 30
Externally publishedYes

Fingerprint

Oxygenation
Blood
Parietal Lobe
blood
Prefrontal Cortex
Brain
brain
Temporal Lobe
Pattern recognition
Time series
ROC Curve
Aging of materials
Cluster Analysis
time series analysis
methodology
prefrontal cortex

ASJC Scopus subject areas

  • Agricultural and Biological Sciences(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Medicine(all)

Cite this

Wee, C. Y., Yap, P. T., Denny, K., Browndyke, J. N., Potter, G. G., Welsh-Bohmer, K. A., ... Shen, D. (2012). Resting-state multi-spectrum functional connectivity networks for identification of MCI patients. PLoS One, 7(5), [e37828]. https://doi.org/10.1371/journal.pone.0037828

Resting-state multi-spectrum functional connectivity networks for identification of MCI patients. / Wee, Chong Yaw; Yap, Pew Thian; Denny, Kevin; Browndyke, Jeffrey N.; Potter, Guy G.; Welsh-Bohmer, Kathleen A.; Wang, Lihong; Shen, Dinggang.

In: PLoS One, Vol. 7, No. 5, e37828, 30.05.2012.

Research output: Contribution to journalArticle

Wee, CY, Yap, PT, Denny, K, Browndyke, JN, Potter, GG, Welsh-Bohmer, KA, Wang, L & Shen, D 2012, 'Resting-state multi-spectrum functional connectivity networks for identification of MCI patients', PLoS One, vol. 7, no. 5, e37828. https://doi.org/10.1371/journal.pone.0037828
Wee CY, Yap PT, Denny K, Browndyke JN, Potter GG, Welsh-Bohmer KA et al. Resting-state multi-spectrum functional connectivity networks for identification of MCI patients. PLoS One. 2012 May 30;7(5). e37828. https://doi.org/10.1371/journal.pone.0037828
Wee, Chong Yaw ; Yap, Pew Thian ; Denny, Kevin ; Browndyke, Jeffrey N. ; Potter, Guy G. ; Welsh-Bohmer, Kathleen A. ; Wang, Lihong ; Shen, Dinggang. / Resting-state multi-spectrum functional connectivity networks for identification of MCI patients. In: PLoS One. 2012 ; Vol. 7, No. 5.
@article{3a0611769e824836844d1daa040ec3da,
title = "Resting-state multi-spectrum functional connectivity networks for identification of MCI patients",
abstract = "In this paper, a high-dimensional pattern classification framework, based on functional associations between brain regions during resting-state, is proposed to accurately identify MCI individuals from subjects who experience normal aging. The proposed technique employs multi-spectrum networks to characterize the complex yet subtle blood oxygenation level dependent (BOLD) signal changes caused by pathological attacks. The utilization of multi-spectrum networks in identifying MCI individuals is motivated by the inherent frequency-specific properties of BOLD spectrum. It is believed that frequency specific information extracted from different spectra may delineate the complex yet subtle variations of BOLD signals more effectively. In the proposed technique, regional mean time series of each region-of-interest (ROI) is band-pass filtered (0:025 ≤ f ≤ 0:100 Hz) before it is decomposed into five frequency sub-bands. Five connectivity networks are constructed, one from each frequency sub-band. Clustering coefficient of each ROI in relation to the other ROIs are extracted as features for classification. Classification accuracy was evaluated via leave-one-out cross-validation to ensure generalization of performance. The classification accuracy obtained by this approach is 86.5{\%}, which is an increase of at least 18.9{\%} from the conventional full-spectrum methods. A cross-validation estimation of the generalization performance shows an area of 0.863 under the receiver operating characteristic (ROC) curve, indicating good diagnostic power. It was also found that, based on the selected features, portions of the prefrontal cortex, orbitofrontal cortex, temporal lobe, and parietal lobe regions provided the most discriminant information for classification, in line with results reported in previous studies. Analysis on individual frequency sub-bands demonstrated that different sub-bands contribute differently to classification, providing extra evidence regarding frequency-specific distribution of BOLD signals. Our MCI classification framework, which allows accurate early detection of functional brain abnormalities, makes an important positive contribution to the treatment management of potential AD patients.",
author = "Wee, {Chong Yaw} and Yap, {Pew Thian} and Kevin Denny and Browndyke, {Jeffrey N.} and Potter, {Guy G.} and Welsh-Bohmer, {Kathleen A.} and Lihong Wang and Dinggang Shen",
year = "2012",
month = "5",
day = "30",
doi = "10.1371/journal.pone.0037828",
language = "English",
volume = "7",
journal = "PLoS One",
issn = "1932-6203",
publisher = "Public Library of Science",
number = "5",

}

TY - JOUR

T1 - Resting-state multi-spectrum functional connectivity networks for identification of MCI patients

AU - Wee, Chong Yaw

AU - Yap, Pew Thian

AU - Denny, Kevin

AU - Browndyke, Jeffrey N.

AU - Potter, Guy G.

AU - Welsh-Bohmer, Kathleen A.

AU - Wang, Lihong

AU - Shen, Dinggang

PY - 2012/5/30

Y1 - 2012/5/30

N2 - In this paper, a high-dimensional pattern classification framework, based on functional associations between brain regions during resting-state, is proposed to accurately identify MCI individuals from subjects who experience normal aging. The proposed technique employs multi-spectrum networks to characterize the complex yet subtle blood oxygenation level dependent (BOLD) signal changes caused by pathological attacks. The utilization of multi-spectrum networks in identifying MCI individuals is motivated by the inherent frequency-specific properties of BOLD spectrum. It is believed that frequency specific information extracted from different spectra may delineate the complex yet subtle variations of BOLD signals more effectively. In the proposed technique, regional mean time series of each region-of-interest (ROI) is band-pass filtered (0:025 ≤ f ≤ 0:100 Hz) before it is decomposed into five frequency sub-bands. Five connectivity networks are constructed, one from each frequency sub-band. Clustering coefficient of each ROI in relation to the other ROIs are extracted as features for classification. Classification accuracy was evaluated via leave-one-out cross-validation to ensure generalization of performance. The classification accuracy obtained by this approach is 86.5%, which is an increase of at least 18.9% from the conventional full-spectrum methods. A cross-validation estimation of the generalization performance shows an area of 0.863 under the receiver operating characteristic (ROC) curve, indicating good diagnostic power. It was also found that, based on the selected features, portions of the prefrontal cortex, orbitofrontal cortex, temporal lobe, and parietal lobe regions provided the most discriminant information for classification, in line with results reported in previous studies. Analysis on individual frequency sub-bands demonstrated that different sub-bands contribute differently to classification, providing extra evidence regarding frequency-specific distribution of BOLD signals. Our MCI classification framework, which allows accurate early detection of functional brain abnormalities, makes an important positive contribution to the treatment management of potential AD patients.

AB - In this paper, a high-dimensional pattern classification framework, based on functional associations between brain regions during resting-state, is proposed to accurately identify MCI individuals from subjects who experience normal aging. The proposed technique employs multi-spectrum networks to characterize the complex yet subtle blood oxygenation level dependent (BOLD) signal changes caused by pathological attacks. The utilization of multi-spectrum networks in identifying MCI individuals is motivated by the inherent frequency-specific properties of BOLD spectrum. It is believed that frequency specific information extracted from different spectra may delineate the complex yet subtle variations of BOLD signals more effectively. In the proposed technique, regional mean time series of each region-of-interest (ROI) is band-pass filtered (0:025 ≤ f ≤ 0:100 Hz) before it is decomposed into five frequency sub-bands. Five connectivity networks are constructed, one from each frequency sub-band. Clustering coefficient of each ROI in relation to the other ROIs are extracted as features for classification. Classification accuracy was evaluated via leave-one-out cross-validation to ensure generalization of performance. The classification accuracy obtained by this approach is 86.5%, which is an increase of at least 18.9% from the conventional full-spectrum methods. A cross-validation estimation of the generalization performance shows an area of 0.863 under the receiver operating characteristic (ROC) curve, indicating good diagnostic power. It was also found that, based on the selected features, portions of the prefrontal cortex, orbitofrontal cortex, temporal lobe, and parietal lobe regions provided the most discriminant information for classification, in line with results reported in previous studies. Analysis on individual frequency sub-bands demonstrated that different sub-bands contribute differently to classification, providing extra evidence regarding frequency-specific distribution of BOLD signals. Our MCI classification framework, which allows accurate early detection of functional brain abnormalities, makes an important positive contribution to the treatment management of potential AD patients.

UR - http://www.scopus.com/inward/record.url?scp=84861644119&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84861644119&partnerID=8YFLogxK

U2 - 10.1371/journal.pone.0037828

DO - 10.1371/journal.pone.0037828

M3 - Article

C2 - 22666397

AN - SCOPUS:84861644119

VL - 7

JO - PLoS One

JF - PLoS One

SN - 1932-6203

IS - 5

M1 - e37828

ER -