Group-constrained sparse fMRI connectivity modeling for mild cognitive impairment identification

Chong Yaw Wee, Pew Thian Yap, Daoqiang Zhang, Lihong Wang, Dinggang Shen

Research output: Contribution to journalArticle

62 Citations (Scopus)

Abstract

Emergence of advanced network analysis techniques utilizing resting-state functional magnetic resonance imaging (R-fMRI) has enabled a more comprehensive understanding of neurological disorders at a whole-brain level. However, inferring brain connectivity from R-fMRI is a challenging task, particularly when the ultimate goal is to achieve good control-patient classification performance, owing to perplexing noise effects, curse of dimensionality, and inter-subject variability. Incorporating sparsity into connectivity modeling may be a possible solution to partially remedy this problem since most biological networks are intrinsically sparse. Nevertheless, sparsity constraint, when applied at an individual level, will inevitably cause inter-subject variability and hence degrade classification performance. To this end, we formulate the R-fMRI time series of each region of interest (ROI) as a linear representation of time series of other ROIs to infer sparse connectivity networks that are topologically identical across individuals. This formulation allows simultaneous selection of a common set of ROIs across subjects so that their linear combination is best in estimating the time series of the considered ROI. Specifically, l1-norm is imposed on each subject to filter out spurious or insignificant connections to produce sparse networks. A group-constraint is hence imposed via multi-task learning using a l 2-norm to encourage consistent non-zero connections across subjects. This group-constraint is crucial since the network topology is identical for all subjects while still preserving individual information via different connectivity values. We validated the proposed modeling in mild cognitive impairment identification and promising results achieved demonstrate its superiority in disease characterization, particularly greater sensitivity to early stage brain pathologies. The inferred group-constrained sparse network is found to be biologically plausible and is highly associated with the disease-associated anatomical anomalies. Furthermore, our proposed approach achieved similar classification performance when finer atlas was used to parcellate the brain space.

Original languageEnglish
Pages (from-to)641-656
Number of pages16
JournalBrain Structure and Function
Volume219
Issue number2
DOIs
Publication statusPublished - 2014 Jan 1
Externally publishedYes

Fingerprint

Magnetic Resonance Imaging
Brain
Atlases
Nervous System Diseases
Noise
Learning
Pathology
Cognitive Dysfunction
Identification (Psychology)

ASJC Scopus subject areas

  • Anatomy
  • Histology
  • Neuroscience(all)

Cite this

Group-constrained sparse fMRI connectivity modeling for mild cognitive impairment identification. / Wee, Chong Yaw; Yap, Pew Thian; Zhang, Daoqiang; Wang, Lihong; Shen, Dinggang.

In: Brain Structure and Function, Vol. 219, No. 2, 01.01.2014, p. 641-656.

Research output: Contribution to journalArticle

Wee, Chong Yaw ; Yap, Pew Thian ; Zhang, Daoqiang ; Wang, Lihong ; Shen, Dinggang. / Group-constrained sparse fMRI connectivity modeling for mild cognitive impairment identification. In: Brain Structure and Function. 2014 ; Vol. 219, No. 2. pp. 641-656.
@article{1a9edb746d4e4a118bd2d17307581fe9,
title = "Group-constrained sparse fMRI connectivity modeling for mild cognitive impairment identification",
abstract = "Emergence of advanced network analysis techniques utilizing resting-state functional magnetic resonance imaging (R-fMRI) has enabled a more comprehensive understanding of neurological disorders at a whole-brain level. However, inferring brain connectivity from R-fMRI is a challenging task, particularly when the ultimate goal is to achieve good control-patient classification performance, owing to perplexing noise effects, curse of dimensionality, and inter-subject variability. Incorporating sparsity into connectivity modeling may be a possible solution to partially remedy this problem since most biological networks are intrinsically sparse. Nevertheless, sparsity constraint, when applied at an individual level, will inevitably cause inter-subject variability and hence degrade classification performance. To this end, we formulate the R-fMRI time series of each region of interest (ROI) as a linear representation of time series of other ROIs to infer sparse connectivity networks that are topologically identical across individuals. This formulation allows simultaneous selection of a common set of ROIs across subjects so that their linear combination is best in estimating the time series of the considered ROI. Specifically, l1-norm is imposed on each subject to filter out spurious or insignificant connections to produce sparse networks. A group-constraint is hence imposed via multi-task learning using a l 2-norm to encourage consistent non-zero connections across subjects. This group-constraint is crucial since the network topology is identical for all subjects while still preserving individual information via different connectivity values. We validated the proposed modeling in mild cognitive impairment identification and promising results achieved demonstrate its superiority in disease characterization, particularly greater sensitivity to early stage brain pathologies. The inferred group-constrained sparse network is found to be biologically plausible and is highly associated with the disease-associated anatomical anomalies. Furthermore, our proposed approach achieved similar classification performance when finer atlas was used to parcellate the brain space.",
keywords = "Group-constrained sparse modeling, Inter-subject variability, Mild cognitive impairment (MCI), Multi-task learning, Resting-state fMRI, Sparse linear regression",
author = "Wee, {Chong Yaw} and Yap, {Pew Thian} and Daoqiang Zhang and Lihong Wang and Dinggang Shen",
year = "2014",
month = "1",
day = "1",
doi = "10.1007/s00429-013-0524-8",
language = "English",
volume = "219",
pages = "641--656",
journal = "Anatomische Hefte",
issn = "0177-5154",
publisher = "Springer Verlag",
number = "2",

}

TY - JOUR

T1 - Group-constrained sparse fMRI connectivity modeling for mild cognitive impairment identification

AU - Wee, Chong Yaw

AU - Yap, Pew Thian

AU - Zhang, Daoqiang

AU - Wang, Lihong

AU - Shen, Dinggang

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Emergence of advanced network analysis techniques utilizing resting-state functional magnetic resonance imaging (R-fMRI) has enabled a more comprehensive understanding of neurological disorders at a whole-brain level. However, inferring brain connectivity from R-fMRI is a challenging task, particularly when the ultimate goal is to achieve good control-patient classification performance, owing to perplexing noise effects, curse of dimensionality, and inter-subject variability. Incorporating sparsity into connectivity modeling may be a possible solution to partially remedy this problem since most biological networks are intrinsically sparse. Nevertheless, sparsity constraint, when applied at an individual level, will inevitably cause inter-subject variability and hence degrade classification performance. To this end, we formulate the R-fMRI time series of each region of interest (ROI) as a linear representation of time series of other ROIs to infer sparse connectivity networks that are topologically identical across individuals. This formulation allows simultaneous selection of a common set of ROIs across subjects so that their linear combination is best in estimating the time series of the considered ROI. Specifically, l1-norm is imposed on each subject to filter out spurious or insignificant connections to produce sparse networks. A group-constraint is hence imposed via multi-task learning using a l 2-norm to encourage consistent non-zero connections across subjects. This group-constraint is crucial since the network topology is identical for all subjects while still preserving individual information via different connectivity values. We validated the proposed modeling in mild cognitive impairment identification and promising results achieved demonstrate its superiority in disease characterization, particularly greater sensitivity to early stage brain pathologies. The inferred group-constrained sparse network is found to be biologically plausible and is highly associated with the disease-associated anatomical anomalies. Furthermore, our proposed approach achieved similar classification performance when finer atlas was used to parcellate the brain space.

AB - Emergence of advanced network analysis techniques utilizing resting-state functional magnetic resonance imaging (R-fMRI) has enabled a more comprehensive understanding of neurological disorders at a whole-brain level. However, inferring brain connectivity from R-fMRI is a challenging task, particularly when the ultimate goal is to achieve good control-patient classification performance, owing to perplexing noise effects, curse of dimensionality, and inter-subject variability. Incorporating sparsity into connectivity modeling may be a possible solution to partially remedy this problem since most biological networks are intrinsically sparse. Nevertheless, sparsity constraint, when applied at an individual level, will inevitably cause inter-subject variability and hence degrade classification performance. To this end, we formulate the R-fMRI time series of each region of interest (ROI) as a linear representation of time series of other ROIs to infer sparse connectivity networks that are topologically identical across individuals. This formulation allows simultaneous selection of a common set of ROIs across subjects so that their linear combination is best in estimating the time series of the considered ROI. Specifically, l1-norm is imposed on each subject to filter out spurious or insignificant connections to produce sparse networks. A group-constraint is hence imposed via multi-task learning using a l 2-norm to encourage consistent non-zero connections across subjects. This group-constraint is crucial since the network topology is identical for all subjects while still preserving individual information via different connectivity values. We validated the proposed modeling in mild cognitive impairment identification and promising results achieved demonstrate its superiority in disease characterization, particularly greater sensitivity to early stage brain pathologies. The inferred group-constrained sparse network is found to be biologically plausible and is highly associated with the disease-associated anatomical anomalies. Furthermore, our proposed approach achieved similar classification performance when finer atlas was used to parcellate the brain space.

KW - Group-constrained sparse modeling

KW - Inter-subject variability

KW - Mild cognitive impairment (MCI)

KW - Multi-task learning

KW - Resting-state fMRI

KW - Sparse linear regression

UR - http://www.scopus.com/inward/record.url?scp=84896048486&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84896048486&partnerID=8YFLogxK

U2 - 10.1007/s00429-013-0524-8

DO - 10.1007/s00429-013-0524-8

M3 - Article

VL - 219

SP - 641

EP - 656

JO - Anatomische Hefte

JF - Anatomische Hefte

SN - 0177-5154

IS - 2

ER -