Relationship Induced Multi-Template Learning for Diagnosis of Alzheimer's Disease and Mild Cognitive Impairment

Mingxia Liu, Daoqiang Zhang, Dinggang Shen

Research output: Contribution to journalArticle

67 Citations (Scopus)

Abstract

As shown in the literature, methods based on multiple templates usually achieve better performance, compared with those using only a single template for processing medical images. However, most existing multi-template based methods simply average or concatenate multiple sets of features extracted from different templates, which potentially ignores important structural information contained in the multi-template data. Accordingly, in this paper, we propose a novel relationship induced multi-template learning method for automatic diagnosis of Alzheimer's disease (AD) and its prodromal stage, i.e., mild cognitive impairment (MCI), by explicitly modeling structural information in the multi-template data. Specifically, we first nonlinearly register each brain's magnetic resonance (MR) image separately onto multiple pre-selected templates, and then extract multiple sets of features for this MR image. Next, we develop a novel feature selection algorithm by introducing two regularization terms to model the relationships among templates and among individual subjects. Using these selected features corresponding to multiple templates, we then construct multiple support vector machine (SVM) classifiers. Finally, an ensemble classification is used to combine outputs of all SVM classifiers, for achieving the final result. We evaluate our proposed method on 459 subjects from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database, including 97 AD patients, 128 normal controls (NC), 117 progressive MCI (pMCI) patients, and 117 stable MCI (sMCI) patients. The experimental results demonstrate promising classification performance, compared with several state-of-the-art methods for multi-template based AD/MCI classification.

Original languageEnglish
Article number7372483
Pages (from-to)1463-1474
Number of pages12
JournalIEEE Transactions on Medical Imaging
Volume35
Issue number6
DOIs
Publication statusPublished - 2016 Jun 1

Fingerprint

Alzheimer Disease
Learning
Magnetic resonance
Support vector machines
Classifiers
Medical image processing
Neuroimaging
Magnetic Resonance Spectroscopy
Prodromal Symptoms
Feature extraction
Brain
Cognitive Dysfunction
Databases
Support Vector Machine

Keywords

  • Alzheimer's disease
  • ensemble classification
  • multi-template representation
  • sparse feature selection

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Science Applications
  • Radiological and Ultrasound Technology
  • Software

Cite this

Relationship Induced Multi-Template Learning for Diagnosis of Alzheimer's Disease and Mild Cognitive Impairment. / Liu, Mingxia; Zhang, Daoqiang; Shen, Dinggang.

In: IEEE Transactions on Medical Imaging, Vol. 35, No. 6, 7372483, 01.06.2016, p. 1463-1474.

Research output: Contribution to journalArticle

@article{5ca30b84b1bf421ebef01a3f13baf2b1,
title = "Relationship Induced Multi-Template Learning for Diagnosis of Alzheimer's Disease and Mild Cognitive Impairment",
abstract = "As shown in the literature, methods based on multiple templates usually achieve better performance, compared with those using only a single template for processing medical images. However, most existing multi-template based methods simply average or concatenate multiple sets of features extracted from different templates, which potentially ignores important structural information contained in the multi-template data. Accordingly, in this paper, we propose a novel relationship induced multi-template learning method for automatic diagnosis of Alzheimer's disease (AD) and its prodromal stage, i.e., mild cognitive impairment (MCI), by explicitly modeling structural information in the multi-template data. Specifically, we first nonlinearly register each brain's magnetic resonance (MR) image separately onto multiple pre-selected templates, and then extract multiple sets of features for this MR image. Next, we develop a novel feature selection algorithm by introducing two regularization terms to model the relationships among templates and among individual subjects. Using these selected features corresponding to multiple templates, we then construct multiple support vector machine (SVM) classifiers. Finally, an ensemble classification is used to combine outputs of all SVM classifiers, for achieving the final result. We evaluate our proposed method on 459 subjects from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database, including 97 AD patients, 128 normal controls (NC), 117 progressive MCI (pMCI) patients, and 117 stable MCI (sMCI) patients. The experimental results demonstrate promising classification performance, compared with several state-of-the-art methods for multi-template based AD/MCI classification.",
keywords = "Alzheimer's disease, ensemble classification, multi-template representation, sparse feature selection",
author = "Mingxia Liu and Daoqiang Zhang and Dinggang Shen",
year = "2016",
month = "6",
day = "1",
doi = "10.1109/TMI.2016.2515021",
language = "English",
volume = "35",
pages = "1463--1474",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "6",

}

TY - JOUR

T1 - Relationship Induced Multi-Template Learning for Diagnosis of Alzheimer's Disease and Mild Cognitive Impairment

AU - Liu, Mingxia

AU - Zhang, Daoqiang

AU - Shen, Dinggang

PY - 2016/6/1

Y1 - 2016/6/1

N2 - As shown in the literature, methods based on multiple templates usually achieve better performance, compared with those using only a single template for processing medical images. However, most existing multi-template based methods simply average or concatenate multiple sets of features extracted from different templates, which potentially ignores important structural information contained in the multi-template data. Accordingly, in this paper, we propose a novel relationship induced multi-template learning method for automatic diagnosis of Alzheimer's disease (AD) and its prodromal stage, i.e., mild cognitive impairment (MCI), by explicitly modeling structural information in the multi-template data. Specifically, we first nonlinearly register each brain's magnetic resonance (MR) image separately onto multiple pre-selected templates, and then extract multiple sets of features for this MR image. Next, we develop a novel feature selection algorithm by introducing two regularization terms to model the relationships among templates and among individual subjects. Using these selected features corresponding to multiple templates, we then construct multiple support vector machine (SVM) classifiers. Finally, an ensemble classification is used to combine outputs of all SVM classifiers, for achieving the final result. We evaluate our proposed method on 459 subjects from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database, including 97 AD patients, 128 normal controls (NC), 117 progressive MCI (pMCI) patients, and 117 stable MCI (sMCI) patients. The experimental results demonstrate promising classification performance, compared with several state-of-the-art methods for multi-template based AD/MCI classification.

AB - As shown in the literature, methods based on multiple templates usually achieve better performance, compared with those using only a single template for processing medical images. However, most existing multi-template based methods simply average or concatenate multiple sets of features extracted from different templates, which potentially ignores important structural information contained in the multi-template data. Accordingly, in this paper, we propose a novel relationship induced multi-template learning method for automatic diagnosis of Alzheimer's disease (AD) and its prodromal stage, i.e., mild cognitive impairment (MCI), by explicitly modeling structural information in the multi-template data. Specifically, we first nonlinearly register each brain's magnetic resonance (MR) image separately onto multiple pre-selected templates, and then extract multiple sets of features for this MR image. Next, we develop a novel feature selection algorithm by introducing two regularization terms to model the relationships among templates and among individual subjects. Using these selected features corresponding to multiple templates, we then construct multiple support vector machine (SVM) classifiers. Finally, an ensemble classification is used to combine outputs of all SVM classifiers, for achieving the final result. We evaluate our proposed method on 459 subjects from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database, including 97 AD patients, 128 normal controls (NC), 117 progressive MCI (pMCI) patients, and 117 stable MCI (sMCI) patients. The experimental results demonstrate promising classification performance, compared with several state-of-the-art methods for multi-template based AD/MCI classification.

KW - Alzheimer's disease

KW - ensemble classification

KW - multi-template representation

KW - sparse feature selection

UR - http://www.scopus.com/inward/record.url?scp=84974687816&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84974687816&partnerID=8YFLogxK

U2 - 10.1109/TMI.2016.2515021

DO - 10.1109/TMI.2016.2515021

M3 - Article

C2 - 26742127

AN - SCOPUS:84974687816

VL - 35

SP - 1463

EP - 1474

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 6

M1 - 7372483

ER -