Sub-network Kernels for Measuring Similarity of Brain Connectivity Networks in Disease Diagnosis

Biao Jie, Mingxia Liu, Daoqiang Zhang, Dinggang Shen

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

As a simple representation of interactions among distributed brain regions, brain networks have been widely applied to automated diagnosis of brain diseases, such as Alzheimer’s disease (AD) and its early stage, i.e., mild cognitive impairment (MCI). In brain network analysis, a challenging task is how to measure the similarity between a pair of networks. Although many graph kernels (i.e., kernels defined on graphs) have been proposed for measuring the topological similarity of a pair of brain networks, most of them are defined using general graphs, thus ignoring the uniqueness of each node in brain networks. That is, each node in a brain network denotes a particular brain region, which is a specific characteristics of brain networks. Accordingly, in this paper, we construct a novel sub-network kernel for measuring the similarity between a pair of brain networks and then apply it to brain disease classification. Different from current graph kernels, our proposed sub-network kernel not only takes into account the inherent characteristic of brain networks, but also captures multi-level (from local to global) topological properties of nodes in brain networks, which are essential for defining the similarity measure of brain networks. To validate the efficacy of our method, we perform extensive experiments on subjects with baseline functional magnetic resonance imaging (fMRI) data obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database. Experimental results demonstrate that the proposed method outperforms several stateof- the-art graph-based methods in MCI classification.

Original languageEnglish
JournalIEEE Transactions on Image Processing
DOIs
Publication statusAccepted/In press - 2018 Jan 29

Fingerprint

Brain
Brain Diseases
Neuroimaging
Art
Electric network analysis
Magnetic Resonance Imaging
Databases

Keywords

  • Alzheimer’s disease (AD)
  • Brain
  • brain network
  • classification
  • Dementia
  • Functional magnetic resonance imaging
  • Graph kernel
  • Graph theory
  • Kernel
  • Length measurement
  • mild cognitive impairment (MCI)

ASJC Scopus subject areas

  • Software
  • Computer Graphics and Computer-Aided Design

Cite this

Sub-network Kernels for Measuring Similarity of Brain Connectivity Networks in Disease Diagnosis. / Jie, Biao; Liu, Mingxia; Zhang, Daoqiang; Shen, Dinggang.

In: IEEE Transactions on Image Processing, 29.01.2018.

Research output: Contribution to journalArticle

@article{f2f1aac50d604e709378549b813d27a4,
title = "Sub-network Kernels for Measuring Similarity of Brain Connectivity Networks in Disease Diagnosis",
abstract = "As a simple representation of interactions among distributed brain regions, brain networks have been widely applied to automated diagnosis of brain diseases, such as Alzheimer’s disease (AD) and its early stage, i.e., mild cognitive impairment (MCI). In brain network analysis, a challenging task is how to measure the similarity between a pair of networks. Although many graph kernels (i.e., kernels defined on graphs) have been proposed for measuring the topological similarity of a pair of brain networks, most of them are defined using general graphs, thus ignoring the uniqueness of each node in brain networks. That is, each node in a brain network denotes a particular brain region, which is a specific characteristics of brain networks. Accordingly, in this paper, we construct a novel sub-network kernel for measuring the similarity between a pair of brain networks and then apply it to brain disease classification. Different from current graph kernels, our proposed sub-network kernel not only takes into account the inherent characteristic of brain networks, but also captures multi-level (from local to global) topological properties of nodes in brain networks, which are essential for defining the similarity measure of brain networks. To validate the efficacy of our method, we perform extensive experiments on subjects with baseline functional magnetic resonance imaging (fMRI) data obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database. Experimental results demonstrate that the proposed method outperforms several stateof- the-art graph-based methods in MCI classification.",
keywords = "Alzheimer’s disease (AD), Brain, brain network, classification, Dementia, Functional magnetic resonance imaging, Graph kernel, Graph theory, Kernel, Length measurement, mild cognitive impairment (MCI)",
author = "Biao Jie and Mingxia Liu and Daoqiang Zhang and Dinggang Shen",
year = "2018",
month = "1",
day = "29",
doi = "10.1109/TIP.2018.2799706",
language = "English",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - Sub-network Kernels for Measuring Similarity of Brain Connectivity Networks in Disease Diagnosis

AU - Jie, Biao

AU - Liu, Mingxia

AU - Zhang, Daoqiang

AU - Shen, Dinggang

PY - 2018/1/29

Y1 - 2018/1/29

N2 - As a simple representation of interactions among distributed brain regions, brain networks have been widely applied to automated diagnosis of brain diseases, such as Alzheimer’s disease (AD) and its early stage, i.e., mild cognitive impairment (MCI). In brain network analysis, a challenging task is how to measure the similarity between a pair of networks. Although many graph kernels (i.e., kernels defined on graphs) have been proposed for measuring the topological similarity of a pair of brain networks, most of them are defined using general graphs, thus ignoring the uniqueness of each node in brain networks. That is, each node in a brain network denotes a particular brain region, which is a specific characteristics of brain networks. Accordingly, in this paper, we construct a novel sub-network kernel for measuring the similarity between a pair of brain networks and then apply it to brain disease classification. Different from current graph kernels, our proposed sub-network kernel not only takes into account the inherent characteristic of brain networks, but also captures multi-level (from local to global) topological properties of nodes in brain networks, which are essential for defining the similarity measure of brain networks. To validate the efficacy of our method, we perform extensive experiments on subjects with baseline functional magnetic resonance imaging (fMRI) data obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database. Experimental results demonstrate that the proposed method outperforms several stateof- the-art graph-based methods in MCI classification.

AB - As a simple representation of interactions among distributed brain regions, brain networks have been widely applied to automated diagnosis of brain diseases, such as Alzheimer’s disease (AD) and its early stage, i.e., mild cognitive impairment (MCI). In brain network analysis, a challenging task is how to measure the similarity between a pair of networks. Although many graph kernels (i.e., kernels defined on graphs) have been proposed for measuring the topological similarity of a pair of brain networks, most of them are defined using general graphs, thus ignoring the uniqueness of each node in brain networks. That is, each node in a brain network denotes a particular brain region, which is a specific characteristics of brain networks. Accordingly, in this paper, we construct a novel sub-network kernel for measuring the similarity between a pair of brain networks and then apply it to brain disease classification. Different from current graph kernels, our proposed sub-network kernel not only takes into account the inherent characteristic of brain networks, but also captures multi-level (from local to global) topological properties of nodes in brain networks, which are essential for defining the similarity measure of brain networks. To validate the efficacy of our method, we perform extensive experiments on subjects with baseline functional magnetic resonance imaging (fMRI) data obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database. Experimental results demonstrate that the proposed method outperforms several stateof- the-art graph-based methods in MCI classification.

KW - Alzheimer’s disease (AD)

KW - Brain

KW - brain network

KW - classification

KW - Dementia

KW - Functional magnetic resonance imaging

KW - Graph kernel

KW - Graph theory

KW - Kernel

KW - Length measurement

KW - mild cognitive impairment (MCI)

UR - http://www.scopus.com/inward/record.url?scp=85041391129&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85041391129&partnerID=8YFLogxK

U2 - 10.1109/TIP.2018.2799706

DO - 10.1109/TIP.2018.2799706

M3 - Article

C2 - 29470170

AN - SCOPUS:85041391129

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

ER -