Feasibility study of a multi-criteria decision-making based hierarchical model for multi-modality feature and multi-classifier fusion

Applications in medical prognosis prediction

Qiang He, Xin Li, D. W.Nathan Kim, Xun Jia, Xuejun Gu, Xin Zhen, Linghong Zhou

Research output: Contribution to journalArticle

Abstract

Radiomics has great prospects in terms of tumour grading, diagnosis and prediction of prognosis by analysing multifaceted data from sources such as clinical treatments, medical images, and pathology. However, exploring an effective way to manage miscellaneous clinical information, as well as to select an appropriate classifier for prediction modelling, is still demanding in a practical clinical context. In this study, we propose a multi-criterion decision-making (MCDM) based classifier fusion (MCF) strategy to combine different classifiers within an MCDM framework. A hierarchical predictive scheme (H-MCF) based on the proposed MCF is also investigated to reliably link the multi-modality features and multi-classifiers. Ten public UCI datasets and two clinical datasets were used to validate the proposed MCF and H-MCF. The experimental results showed that H-MCF has superior predictive performance when compared with the traditional fusion strategies and other fusion architectures, thus demonstrating the feasibility of the proposed H-MCF in integrating information from features of diversified modalities and different classifiers.

Original languageEnglish (US)
Pages (from-to)207-219
Number of pages13
JournalInformation Fusion
Volume55
DOIs
StatePublished - Mar 1 2020

Fingerprint

Classifiers
Fusion reactions
Decision making
Pathology
Tumors

Keywords

  • Classifier fusion
  • Multi-classifier
  • Multi-modality
  • Multiple criteria decision making
  • Radiomics

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Information Systems
  • Hardware and Architecture

Cite this

Feasibility study of a multi-criteria decision-making based hierarchical model for multi-modality feature and multi-classifier fusion : Applications in medical prognosis prediction. / He, Qiang; Li, Xin; Kim, D. W.Nathan; Jia, Xun; Gu, Xuejun; Zhen, Xin; Zhou, Linghong.

In: Information Fusion, Vol. 55, 01.03.2020, p. 207-219.

Research output: Contribution to journalArticle

@article{b123986ebdf849d1afa6c9c8b8489106,
title = "Feasibility study of a multi-criteria decision-making based hierarchical model for multi-modality feature and multi-classifier fusion: Applications in medical prognosis prediction",
abstract = "Radiomics has great prospects in terms of tumour grading, diagnosis and prediction of prognosis by analysing multifaceted data from sources such as clinical treatments, medical images, and pathology. However, exploring an effective way to manage miscellaneous clinical information, as well as to select an appropriate classifier for prediction modelling, is still demanding in a practical clinical context. In this study, we propose a multi-criterion decision-making (MCDM) based classifier fusion (MCF) strategy to combine different classifiers within an MCDM framework. A hierarchical predictive scheme (H-MCF) based on the proposed MCF is also investigated to reliably link the multi-modality features and multi-classifiers. Ten public UCI datasets and two clinical datasets were used to validate the proposed MCF and H-MCF. The experimental results showed that H-MCF has superior predictive performance when compared with the traditional fusion strategies and other fusion architectures, thus demonstrating the feasibility of the proposed H-MCF in integrating information from features of diversified modalities and different classifiers.",
keywords = "Classifier fusion, Multi-classifier, Multi-modality, Multiple criteria decision making, Radiomics",
author = "Qiang He and Xin Li and Kim, {D. W.Nathan} and Xun Jia and Xuejun Gu and Xin Zhen and Linghong Zhou",
year = "2020",
month = "3",
day = "1",
doi = "10.1016/j.inffus.2019.09.001",
language = "English (US)",
volume = "55",
pages = "207--219",
journal = "Information Fusion",
issn = "1566-2535",
publisher = "Elsevier",

}

TY - JOUR

T1 - Feasibility study of a multi-criteria decision-making based hierarchical model for multi-modality feature and multi-classifier fusion

T2 - Applications in medical prognosis prediction

AU - He, Qiang

AU - Li, Xin

AU - Kim, D. W.Nathan

AU - Jia, Xun

AU - Gu, Xuejun

AU - Zhen, Xin

AU - Zhou, Linghong

PY - 2020/3/1

Y1 - 2020/3/1

N2 - Radiomics has great prospects in terms of tumour grading, diagnosis and prediction of prognosis by analysing multifaceted data from sources such as clinical treatments, medical images, and pathology. However, exploring an effective way to manage miscellaneous clinical information, as well as to select an appropriate classifier for prediction modelling, is still demanding in a practical clinical context. In this study, we propose a multi-criterion decision-making (MCDM) based classifier fusion (MCF) strategy to combine different classifiers within an MCDM framework. A hierarchical predictive scheme (H-MCF) based on the proposed MCF is also investigated to reliably link the multi-modality features and multi-classifiers. Ten public UCI datasets and two clinical datasets were used to validate the proposed MCF and H-MCF. The experimental results showed that H-MCF has superior predictive performance when compared with the traditional fusion strategies and other fusion architectures, thus demonstrating the feasibility of the proposed H-MCF in integrating information from features of diversified modalities and different classifiers.

AB - Radiomics has great prospects in terms of tumour grading, diagnosis and prediction of prognosis by analysing multifaceted data from sources such as clinical treatments, medical images, and pathology. However, exploring an effective way to manage miscellaneous clinical information, as well as to select an appropriate classifier for prediction modelling, is still demanding in a practical clinical context. In this study, we propose a multi-criterion decision-making (MCDM) based classifier fusion (MCF) strategy to combine different classifiers within an MCDM framework. A hierarchical predictive scheme (H-MCF) based on the proposed MCF is also investigated to reliably link the multi-modality features and multi-classifiers. Ten public UCI datasets and two clinical datasets were used to validate the proposed MCF and H-MCF. The experimental results showed that H-MCF has superior predictive performance when compared with the traditional fusion strategies and other fusion architectures, thus demonstrating the feasibility of the proposed H-MCF in integrating information from features of diversified modalities and different classifiers.

KW - Classifier fusion

KW - Multi-classifier

KW - Multi-modality

KW - Multiple criteria decision making

KW - Radiomics

UR - http://www.scopus.com/inward/record.url?scp=85072231135&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85072231135&partnerID=8YFLogxK

U2 - 10.1016/j.inffus.2019.09.001

DO - 10.1016/j.inffus.2019.09.001

M3 - Article

VL - 55

SP - 207

EP - 219

JO - Information Fusion

JF - Information Fusion

SN - 1566-2535

ER -