A Robust Deep Model for Improved Classification of AD/MCI Patients

Feng Li, Loc Tran, Kim Han Thung, Shuiwang Ji, Dinggang Shen, Jiang Li

Research output: Contribution to journalArticle

74 Citations (Scopus)

Abstract

Accurate classification of Alzheimer's disease (AD) and its prodromal stage, mild cognitive impairment (MCI), plays a critical role in possibly preventing progression of memory impairment and improving quality of life for AD patients. Among many research tasks, it is of a particular interest to identify noninvasive imaging biomarkers for AD diagnosis. In this paper, we present a robust deep learning system to identify different progression stages of AD patients based on MRI and PET scans. We utilized the dropout technique to improve classical deep learning by preventing its weight coadaptation, which is a typical cause of overfitting in deep learning. In addition, we incorporated stability selection, an adaptive learning factor, and a multitask learning strategy into the deep learning framework. We applied the proposed method to the ADNI dataset, and conducted experiments for AD and MCI conversion diagnosis. Experimental results showed that the dropout technique is very effective in AD diagnosis, improving the classification accuracies by 5.9% on average as compared to the classical deep learning methods.

Original languageEnglish
Article number7101222
Pages (from-to)1610-1616
Number of pages7
JournalIEEE Journal of Biomedical and Health Informatics
Volume19
Issue number5
DOIs
Publication statusPublished - 2015 Sep 1

Fingerprint

Alzheimer Disease
Learning
Prodromal Symptoms
Biomarkers
Magnetic resonance imaging
Learning systems
Cognitive Dysfunction
Positron-Emission Tomography
Deep learning
Imaging techniques
Data storage equipment
Quality of Life
Magnetic Resonance Imaging
Weights and Measures
Research
Experiments

Keywords

  • Alzheimer's Disease
  • Deep Learning
  • Early Diagnosis
  • MRI
  • PET

ASJC Scopus subject areas

  • Biotechnology
  • Computer Science Applications
  • Electrical and Electronic Engineering
  • Health Information Management

Cite this

A Robust Deep Model for Improved Classification of AD/MCI Patients. / Li, Feng; Tran, Loc; Thung, Kim Han; Ji, Shuiwang; Shen, Dinggang; Li, Jiang.

In: IEEE Journal of Biomedical and Health Informatics, Vol. 19, No. 5, 7101222, 01.09.2015, p. 1610-1616.

Research output: Contribution to journalArticle

Li, Feng ; Tran, Loc ; Thung, Kim Han ; Ji, Shuiwang ; Shen, Dinggang ; Li, Jiang. / A Robust Deep Model for Improved Classification of AD/MCI Patients. In: IEEE Journal of Biomedical and Health Informatics. 2015 ; Vol. 19, No. 5. pp. 1610-1616.
@article{b273d5009a604815a97a5e58a84010ef,
title = "A Robust Deep Model for Improved Classification of AD/MCI Patients",
abstract = "Accurate classification of Alzheimer's disease (AD) and its prodromal stage, mild cognitive impairment (MCI), plays a critical role in possibly preventing progression of memory impairment and improving quality of life for AD patients. Among many research tasks, it is of a particular interest to identify noninvasive imaging biomarkers for AD diagnosis. In this paper, we present a robust deep learning system to identify different progression stages of AD patients based on MRI and PET scans. We utilized the dropout technique to improve classical deep learning by preventing its weight coadaptation, which is a typical cause of overfitting in deep learning. In addition, we incorporated stability selection, an adaptive learning factor, and a multitask learning strategy into the deep learning framework. We applied the proposed method to the ADNI dataset, and conducted experiments for AD and MCI conversion diagnosis. Experimental results showed that the dropout technique is very effective in AD diagnosis, improving the classification accuracies by 5.9{\%} on average as compared to the classical deep learning methods.",
keywords = "Alzheimer's Disease, Deep Learning, Early Diagnosis, MRI, PET",
author = "Feng Li and Loc Tran and Thung, {Kim Han} and Shuiwang Ji and Dinggang Shen and Jiang Li",
year = "2015",
month = "9",
day = "1",
doi = "10.1109/JBHI.2015.2429556",
language = "English",
volume = "19",
pages = "1610--1616",
journal = "IEEE Journal of Biomedical and Health Informatics",
issn = "2168-2194",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "5",

}

TY - JOUR

T1 - A Robust Deep Model for Improved Classification of AD/MCI Patients

AU - Li, Feng

AU - Tran, Loc

AU - Thung, Kim Han

AU - Ji, Shuiwang

AU - Shen, Dinggang

AU - Li, Jiang

PY - 2015/9/1

Y1 - 2015/9/1

N2 - Accurate classification of Alzheimer's disease (AD) and its prodromal stage, mild cognitive impairment (MCI), plays a critical role in possibly preventing progression of memory impairment and improving quality of life for AD patients. Among many research tasks, it is of a particular interest to identify noninvasive imaging biomarkers for AD diagnosis. In this paper, we present a robust deep learning system to identify different progression stages of AD patients based on MRI and PET scans. We utilized the dropout technique to improve classical deep learning by preventing its weight coadaptation, which is a typical cause of overfitting in deep learning. In addition, we incorporated stability selection, an adaptive learning factor, and a multitask learning strategy into the deep learning framework. We applied the proposed method to the ADNI dataset, and conducted experiments for AD and MCI conversion diagnosis. Experimental results showed that the dropout technique is very effective in AD diagnosis, improving the classification accuracies by 5.9% on average as compared to the classical deep learning methods.

AB - Accurate classification of Alzheimer's disease (AD) and its prodromal stage, mild cognitive impairment (MCI), plays a critical role in possibly preventing progression of memory impairment and improving quality of life for AD patients. Among many research tasks, it is of a particular interest to identify noninvasive imaging biomarkers for AD diagnosis. In this paper, we present a robust deep learning system to identify different progression stages of AD patients based on MRI and PET scans. We utilized the dropout technique to improve classical deep learning by preventing its weight coadaptation, which is a typical cause of overfitting in deep learning. In addition, we incorporated stability selection, an adaptive learning factor, and a multitask learning strategy into the deep learning framework. We applied the proposed method to the ADNI dataset, and conducted experiments for AD and MCI conversion diagnosis. Experimental results showed that the dropout technique is very effective in AD diagnosis, improving the classification accuracies by 5.9% on average as compared to the classical deep learning methods.

KW - Alzheimer's Disease

KW - Deep Learning

KW - Early Diagnosis

KW - MRI

KW - PET

UR - http://www.scopus.com/inward/record.url?scp=84940975497&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84940975497&partnerID=8YFLogxK

U2 - 10.1109/JBHI.2015.2429556

DO - 10.1109/JBHI.2015.2429556

M3 - Article

C2 - 25955998

AN - SCOPUS:84940975497

VL - 19

SP - 1610

EP - 1616

JO - IEEE Journal of Biomedical and Health Informatics

JF - IEEE Journal of Biomedical and Health Informatics

SN - 2168-2194

IS - 5

M1 - 7101222

ER -