Discriminative multi-task feature selection for multi-modality classification of Alzheimer’s disease

Tingting Ye, Chen Zu, Biao Jie, Dinggang Shen, Daoqiang Zhang, Alzheimer’S Disease Neuroimaging Initiative The Alzheimer’S Disease Neuroimaging Initiative

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Recently, multi-task based feature selection methods have been used in multi-modality based classification of Alzheimer’s disease (AD) and its prodromal stage, i.e., mild cognitive impairment (MCI). However, in traditional multi-task feature selection methods, some useful discriminative information among subjects is usually not well mined for further improving the subsequent classification performance. Accordingly, in this paper, we propose a discriminative multi-task feature selection method to select the most discriminative features for multi-modality based classification of AD/MCI. Specifically, for each modality, we train a linear regression model using the corresponding modality of data, and further enforce the group-sparsity regularization on weights of those regression models for joint selection of common features across multiple modalities. Furthermore, we propose a discriminative regularization term based on the intra-class and inter-class Laplacian matrices to better use the discriminative information among subjects. To evaluate our proposed method, we perform extensive experiments on 202 subjects, including 51 AD patients, 99 MCI patients, and 52 healthy controls (HC), from the baseline MRI and FDG-PET image data of the Alzheimer’s Disease Neuroimaging Initiative (ADNI). The experimental results show that our proposed method not only improves the classification performance, but also has potential to discover the disease-related biomarkers useful for diagnosis of disease, along with the comparison to several state-of-the-art methods for multi-modality based AD/MCI classification.

Original languageEnglish
JournalBrain Imaging and Behavior
DOIs
Publication statusAccepted/In press - 2015 Aug 27

Fingerprint

Alzheimer Disease
Linear Models
Prodromal Symptoms
Neuroimaging
Biomarkers
Weights and Measures
Cognitive Dysfunction

Keywords

  • Alzheimer’s disease
  • Discriminative regularization
  • Group-sparsity regularizer
  • Multi-modality based classification
  • Multi-task feature selection

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Behavioral Neuroscience
  • Cellular and Molecular Neuroscience
  • Cognitive Neuroscience
  • Neurology
  • Psychiatry and Mental health
  • Clinical Neurology

Cite this

Ye, T., Zu, C., Jie, B., Shen, D., Zhang, D., & The Alzheimer’S Disease Neuroimaging Initiative, AS. D. N. I. (Accepted/In press). Discriminative multi-task feature selection for multi-modality classification of Alzheimer’s disease. Brain Imaging and Behavior. https://doi.org/10.1007/s11682-015-9437-x

Discriminative multi-task feature selection for multi-modality classification of Alzheimer’s disease. / Ye, Tingting; Zu, Chen; Jie, Biao; Shen, Dinggang; Zhang, Daoqiang; The Alzheimer’S Disease Neuroimaging Initiative, Alzheimer’S Disease Neuroimaging Initiative.

In: Brain Imaging and Behavior, 27.08.2015.

Research output: Contribution to journalArticle

Ye, Tingting ; Zu, Chen ; Jie, Biao ; Shen, Dinggang ; Zhang, Daoqiang ; The Alzheimer’S Disease Neuroimaging Initiative, Alzheimer’S Disease Neuroimaging Initiative. / Discriminative multi-task feature selection for multi-modality classification of Alzheimer’s disease. In: Brain Imaging and Behavior. 2015.
@article{7447e49732b54ed3bc39e4e418387604,
title = "Discriminative multi-task feature selection for multi-modality classification of Alzheimer’s disease",
abstract = "Recently, multi-task based feature selection methods have been used in multi-modality based classification of Alzheimer’s disease (AD) and its prodromal stage, i.e., mild cognitive impairment (MCI). However, in traditional multi-task feature selection methods, some useful discriminative information among subjects is usually not well mined for further improving the subsequent classification performance. Accordingly, in this paper, we propose a discriminative multi-task feature selection method to select the most discriminative features for multi-modality based classification of AD/MCI. Specifically, for each modality, we train a linear regression model using the corresponding modality of data, and further enforce the group-sparsity regularization on weights of those regression models for joint selection of common features across multiple modalities. Furthermore, we propose a discriminative regularization term based on the intra-class and inter-class Laplacian matrices to better use the discriminative information among subjects. To evaluate our proposed method, we perform extensive experiments on 202 subjects, including 51 AD patients, 99 MCI patients, and 52 healthy controls (HC), from the baseline MRI and FDG-PET image data of the Alzheimer’s Disease Neuroimaging Initiative (ADNI). The experimental results show that our proposed method not only improves the classification performance, but also has potential to discover the disease-related biomarkers useful for diagnosis of disease, along with the comparison to several state-of-the-art methods for multi-modality based AD/MCI classification.",
keywords = "Alzheimer’s disease, Discriminative regularization, Group-sparsity regularizer, Multi-modality based classification, Multi-task feature selection",
author = "Tingting Ye and Chen Zu and Biao Jie and Dinggang Shen and Daoqiang Zhang and {The Alzheimer’S Disease Neuroimaging Initiative}, {Alzheimer’S Disease Neuroimaging Initiative}",
year = "2015",
month = "8",
day = "27",
doi = "10.1007/s11682-015-9437-x",
language = "English",
journal = "Brain Imaging and Behavior",
issn = "1931-7557",
publisher = "Springer New York",

}

TY - JOUR

T1 - Discriminative multi-task feature selection for multi-modality classification of Alzheimer’s disease

AU - Ye, Tingting

AU - Zu, Chen

AU - Jie, Biao

AU - Shen, Dinggang

AU - Zhang, Daoqiang

AU - The Alzheimer’S Disease Neuroimaging Initiative, Alzheimer’S Disease Neuroimaging Initiative

PY - 2015/8/27

Y1 - 2015/8/27

N2 - Recently, multi-task based feature selection methods have been used in multi-modality based classification of Alzheimer’s disease (AD) and its prodromal stage, i.e., mild cognitive impairment (MCI). However, in traditional multi-task feature selection methods, some useful discriminative information among subjects is usually not well mined for further improving the subsequent classification performance. Accordingly, in this paper, we propose a discriminative multi-task feature selection method to select the most discriminative features for multi-modality based classification of AD/MCI. Specifically, for each modality, we train a linear regression model using the corresponding modality of data, and further enforce the group-sparsity regularization on weights of those regression models for joint selection of common features across multiple modalities. Furthermore, we propose a discriminative regularization term based on the intra-class and inter-class Laplacian matrices to better use the discriminative information among subjects. To evaluate our proposed method, we perform extensive experiments on 202 subjects, including 51 AD patients, 99 MCI patients, and 52 healthy controls (HC), from the baseline MRI and FDG-PET image data of the Alzheimer’s Disease Neuroimaging Initiative (ADNI). The experimental results show that our proposed method not only improves the classification performance, but also has potential to discover the disease-related biomarkers useful for diagnosis of disease, along with the comparison to several state-of-the-art methods for multi-modality based AD/MCI classification.

AB - Recently, multi-task based feature selection methods have been used in multi-modality based classification of Alzheimer’s disease (AD) and its prodromal stage, i.e., mild cognitive impairment (MCI). However, in traditional multi-task feature selection methods, some useful discriminative information among subjects is usually not well mined for further improving the subsequent classification performance. Accordingly, in this paper, we propose a discriminative multi-task feature selection method to select the most discriminative features for multi-modality based classification of AD/MCI. Specifically, for each modality, we train a linear regression model using the corresponding modality of data, and further enforce the group-sparsity regularization on weights of those regression models for joint selection of common features across multiple modalities. Furthermore, we propose a discriminative regularization term based on the intra-class and inter-class Laplacian matrices to better use the discriminative information among subjects. To evaluate our proposed method, we perform extensive experiments on 202 subjects, including 51 AD patients, 99 MCI patients, and 52 healthy controls (HC), from the baseline MRI and FDG-PET image data of the Alzheimer’s Disease Neuroimaging Initiative (ADNI). The experimental results show that our proposed method not only improves the classification performance, but also has potential to discover the disease-related biomarkers useful for diagnosis of disease, along with the comparison to several state-of-the-art methods for multi-modality based AD/MCI classification.

KW - Alzheimer’s disease

KW - Discriminative regularization

KW - Group-sparsity regularizer

KW - Multi-modality based classification

KW - Multi-task feature selection

UR - http://www.scopus.com/inward/record.url?scp=84940093077&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84940093077&partnerID=8YFLogxK

U2 - 10.1007/s11682-015-9437-x

DO - 10.1007/s11682-015-9437-x

M3 - Article

JO - Brain Imaging and Behavior

JF - Brain Imaging and Behavior

SN - 1931-7557

ER -