Multi-Hypergraph Learning for Incomplete Multi-Modality Data

Mingxia Liu, Yue Gao, Pew Thian Yap, Dinggang Shen

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Multi-modality data convey complementary information that can be used to improve the accuracy of prediction models in disease diagnosis. However, effectively integrating multi-modality data remains a challenging problem especially when the data are incomplete. For instance, more than half of the subjects in the Alzheimer's Disease Neuroimaging Initiative (ADNI) database have no fluorodeoxyglucose positron emission tomography (FDG-PET) and cerebrospinal fluid (CSF) data. Currently, there are two commonly-used strategies to handle the problem of incomplete data: 1) discard samples having missing features, and 2) impute those missing values via specific techniques. In the first case, a significant amount of useful information is lost and, in the second case, additional noise and artifacts might be introduced into the data. Also, previous studies generally focus on the pairwise relationships among subjects, without considering their underlying complex (e.g., high-order) relationships. To address these issues, in this paper, we propose a multi-hypergraph learning (MHL) method for dealing with incomplete multi-modality data. Specifically, we first construct multiple hypergraphs to represent the high-order relationships among subjects by dividing them into several groups according to the availability of their data modalities. A hypergraph regularized transductive learning method is then applied to these groups for automatic diagnosis of brain diseases. Extensive evaluation of the proposed method using all subjects in the baseline ADNI database indicates that our method achieves promising results in AD/MCI classification, compared with state-of-the-art methods.

Original languageEnglish
JournalIEEE Journal of Biomedical and Health Informatics
DOIs
Publication statusAccepted/In press - 2017 Jul 26

Fingerprint

Learning
Neuroimaging
Cerebrospinal fluid
Positron emission tomography
Databases
Brain
Availability
Brain Diseases
Positron-Emission Tomography
Artifacts
Cerebrospinal Fluid
Noise

Keywords

  • Alzheimer's disease
  • classification
  • hypergraph
  • incomplete data
  • Multi-modality

ASJC Scopus subject areas

  • Biotechnology
  • Medicine(all)
  • Computer Science Applications
  • Health Information Management
  • Electrical and Electronic Engineering

Cite this

Multi-Hypergraph Learning for Incomplete Multi-Modality Data. / Liu, Mingxia; Gao, Yue; Yap, Pew Thian; Shen, Dinggang.

In: IEEE Journal of Biomedical and Health Informatics, 26.07.2017.

Research output: Contribution to journalArticle

@article{6412cc23447942fc9ee93df91e725ea3,
title = "Multi-Hypergraph Learning for Incomplete Multi-Modality Data",
abstract = "Multi-modality data convey complementary information that can be used to improve the accuracy of prediction models in disease diagnosis. However, effectively integrating multi-modality data remains a challenging problem especially when the data are incomplete. For instance, more than half of the subjects in the Alzheimer's Disease Neuroimaging Initiative (ADNI) database have no fluorodeoxyglucose positron emission tomography (FDG-PET) and cerebrospinal fluid (CSF) data. Currently, there are two commonly-used strategies to handle the problem of incomplete data: 1) discard samples having missing features, and 2) impute those missing values via specific techniques. In the first case, a significant amount of useful information is lost and, in the second case, additional noise and artifacts might be introduced into the data. Also, previous studies generally focus on the pairwise relationships among subjects, without considering their underlying complex (e.g., high-order) relationships. To address these issues, in this paper, we propose a multi-hypergraph learning (MHL) method for dealing with incomplete multi-modality data. Specifically, we first construct multiple hypergraphs to represent the high-order relationships among subjects by dividing them into several groups according to the availability of their data modalities. A hypergraph regularized transductive learning method is then applied to these groups for automatic diagnosis of brain diseases. Extensive evaluation of the proposed method using all subjects in the baseline ADNI database indicates that our method achieves promising results in AD/MCI classification, compared with state-of-the-art methods.",
keywords = "Alzheimer's disease, classification, hypergraph, incomplete data, Multi-modality",
author = "Mingxia Liu and Yue Gao and Yap, {Pew Thian} and Dinggang Shen",
year = "2017",
month = "7",
day = "26",
doi = "10.1109/JBHI.2017.2732287",
language = "English",
journal = "IEEE Journal of Biomedical and Health Informatics",
issn = "2168-2194",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - Multi-Hypergraph Learning for Incomplete Multi-Modality Data

AU - Liu, Mingxia

AU - Gao, Yue

AU - Yap, Pew Thian

AU - Shen, Dinggang

PY - 2017/7/26

Y1 - 2017/7/26

N2 - Multi-modality data convey complementary information that can be used to improve the accuracy of prediction models in disease diagnosis. However, effectively integrating multi-modality data remains a challenging problem especially when the data are incomplete. For instance, more than half of the subjects in the Alzheimer's Disease Neuroimaging Initiative (ADNI) database have no fluorodeoxyglucose positron emission tomography (FDG-PET) and cerebrospinal fluid (CSF) data. Currently, there are two commonly-used strategies to handle the problem of incomplete data: 1) discard samples having missing features, and 2) impute those missing values via specific techniques. In the first case, a significant amount of useful information is lost and, in the second case, additional noise and artifacts might be introduced into the data. Also, previous studies generally focus on the pairwise relationships among subjects, without considering their underlying complex (e.g., high-order) relationships. To address these issues, in this paper, we propose a multi-hypergraph learning (MHL) method for dealing with incomplete multi-modality data. Specifically, we first construct multiple hypergraphs to represent the high-order relationships among subjects by dividing them into several groups according to the availability of their data modalities. A hypergraph regularized transductive learning method is then applied to these groups for automatic diagnosis of brain diseases. Extensive evaluation of the proposed method using all subjects in the baseline ADNI database indicates that our method achieves promising results in AD/MCI classification, compared with state-of-the-art methods.

AB - Multi-modality data convey complementary information that can be used to improve the accuracy of prediction models in disease diagnosis. However, effectively integrating multi-modality data remains a challenging problem especially when the data are incomplete. For instance, more than half of the subjects in the Alzheimer's Disease Neuroimaging Initiative (ADNI) database have no fluorodeoxyglucose positron emission tomography (FDG-PET) and cerebrospinal fluid (CSF) data. Currently, there are two commonly-used strategies to handle the problem of incomplete data: 1) discard samples having missing features, and 2) impute those missing values via specific techniques. In the first case, a significant amount of useful information is lost and, in the second case, additional noise and artifacts might be introduced into the data. Also, previous studies generally focus on the pairwise relationships among subjects, without considering their underlying complex (e.g., high-order) relationships. To address these issues, in this paper, we propose a multi-hypergraph learning (MHL) method for dealing with incomplete multi-modality data. Specifically, we first construct multiple hypergraphs to represent the high-order relationships among subjects by dividing them into several groups according to the availability of their data modalities. A hypergraph regularized transductive learning method is then applied to these groups for automatic diagnosis of brain diseases. Extensive evaluation of the proposed method using all subjects in the baseline ADNI database indicates that our method achieves promising results in AD/MCI classification, compared with state-of-the-art methods.

KW - Alzheimer's disease

KW - classification

KW - hypergraph

KW - incomplete data

KW - Multi-modality

UR - http://www.scopus.com/inward/record.url?scp=85028942363&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85028942363&partnerID=8YFLogxK

U2 - 10.1109/JBHI.2017.2732287

DO - 10.1109/JBHI.2017.2732287

M3 - Article

C2 - 28749360

AN - SCOPUS:85028942363

JO - IEEE Journal of Biomedical and Health Informatics

JF - IEEE Journal of Biomedical and Health Informatics

SN - 2168-2194

ER -