Cognitive Assessment Prediction in Alzheimer’s Disease by Multi-Layer Multi-Target Regression

Xiaoqian Wang, Xiantong Zhen, Quanzheng Li, Dinggang Shen, Heng Huang

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Accurate and automatic prediction of cognitive assessment from multiple neuroimaging biomarkers is crucial for early detection of Alzheimer’s disease. The major challenges arise from the nonlinear relationship between biomarkers and assessment scores and the inter-correlation among them, which have not yet been well addressed. In this paper, we propose multi-layer multi-target regression (MMR) which enables simultaneously modeling intrinsic inter-target correlations and nonlinear input-output relationships in a general compositional framework. Specifically, by kernelized dictionary learning, the MMR can effectively handle highly nonlinear relationship between biomarkers and assessment scores; by robust low-rank linear learning via matrix elastic nets, the MMR can explicitly encode inter-correlations among multiple assessment scores; moreover, the MMR is flexibly and allows to work with non-smooth ℓ2,1-norm loss function, which enables calibration of multiple targets with disparate noise levels for more robust parameter estimation. The MMR can be efficiently solved by an alternating optimization algorithm via gradient descent with guaranteed convergence. The MMR has been evaluated by extensive experiments on the ADNI database with MRI data, and produced high accuracy surpassing previous regression models, which demonstrates its great effectiveness as a new multi-target regression model for clinical multivariate prediction.

Original languageEnglish
Pages (from-to)1-10
Number of pages10
JournalNeuroinformatics
DOIs
Publication statusAccepted/In press - 2018 May 25
Externally publishedYes

Fingerprint

Biomarkers
Alzheimer Disease
Learning
Neuroimaging
Glossaries
Magnetic resonance imaging
Parameter estimation
Calibration
Noise
Early Diagnosis
Databases
Experiments

Keywords

  • Alzheimer’s disease
  • Calibration
  • Multi-target regression
  • Nonlinear regression
  • Robust low-rank learning

ASJC Scopus subject areas

  • Software
  • Neuroscience(all)
  • Information Systems

Cite this

Cognitive Assessment Prediction in Alzheimer’s Disease by Multi-Layer Multi-Target Regression. / Wang, Xiaoqian; Zhen, Xiantong; Li, Quanzheng; Shen, Dinggang; Huang, Heng.

In: Neuroinformatics, 25.05.2018, p. 1-10.

Research output: Contribution to journalArticle

Wang, Xiaoqian ; Zhen, Xiantong ; Li, Quanzheng ; Shen, Dinggang ; Huang, Heng. / Cognitive Assessment Prediction in Alzheimer’s Disease by Multi-Layer Multi-Target Regression. In: Neuroinformatics. 2018 ; pp. 1-10.
@article{fd4f707eabff480cb9ed2618ae4a1724,
title = "Cognitive Assessment Prediction in Alzheimer’s Disease by Multi-Layer Multi-Target Regression",
abstract = "Accurate and automatic prediction of cognitive assessment from multiple neuroimaging biomarkers is crucial for early detection of Alzheimer’s disease. The major challenges arise from the nonlinear relationship between biomarkers and assessment scores and the inter-correlation among them, which have not yet been well addressed. In this paper, we propose multi-layer multi-target regression (MMR) which enables simultaneously modeling intrinsic inter-target correlations and nonlinear input-output relationships in a general compositional framework. Specifically, by kernelized dictionary learning, the MMR can effectively handle highly nonlinear relationship between biomarkers and assessment scores; by robust low-rank linear learning via matrix elastic nets, the MMR can explicitly encode inter-correlations among multiple assessment scores; moreover, the MMR is flexibly and allows to work with non-smooth ℓ2,1-norm loss function, which enables calibration of multiple targets with disparate noise levels for more robust parameter estimation. The MMR can be efficiently solved by an alternating optimization algorithm via gradient descent with guaranteed convergence. The MMR has been evaluated by extensive experiments on the ADNI database with MRI data, and produced high accuracy surpassing previous regression models, which demonstrates its great effectiveness as a new multi-target regression model for clinical multivariate prediction.",
keywords = "Alzheimer’s disease, Calibration, Multi-target regression, Nonlinear regression, Robust low-rank learning",
author = "Xiaoqian Wang and Xiantong Zhen and Quanzheng Li and Dinggang Shen and Heng Huang",
year = "2018",
month = "5",
day = "25",
doi = "10.1007/s12021-018-9381-1",
language = "English",
pages = "1--10",
journal = "Neuroinformatics",
issn = "1539-2791",
publisher = "Humana Press",

}

TY - JOUR

T1 - Cognitive Assessment Prediction in Alzheimer’s Disease by Multi-Layer Multi-Target Regression

AU - Wang, Xiaoqian

AU - Zhen, Xiantong

AU - Li, Quanzheng

AU - Shen, Dinggang

AU - Huang, Heng

PY - 2018/5/25

Y1 - 2018/5/25

N2 - Accurate and automatic prediction of cognitive assessment from multiple neuroimaging biomarkers is crucial for early detection of Alzheimer’s disease. The major challenges arise from the nonlinear relationship between biomarkers and assessment scores and the inter-correlation among them, which have not yet been well addressed. In this paper, we propose multi-layer multi-target regression (MMR) which enables simultaneously modeling intrinsic inter-target correlations and nonlinear input-output relationships in a general compositional framework. Specifically, by kernelized dictionary learning, the MMR can effectively handle highly nonlinear relationship between biomarkers and assessment scores; by robust low-rank linear learning via matrix elastic nets, the MMR can explicitly encode inter-correlations among multiple assessment scores; moreover, the MMR is flexibly and allows to work with non-smooth ℓ2,1-norm loss function, which enables calibration of multiple targets with disparate noise levels for more robust parameter estimation. The MMR can be efficiently solved by an alternating optimization algorithm via gradient descent with guaranteed convergence. The MMR has been evaluated by extensive experiments on the ADNI database with MRI data, and produced high accuracy surpassing previous regression models, which demonstrates its great effectiveness as a new multi-target regression model for clinical multivariate prediction.

AB - Accurate and automatic prediction of cognitive assessment from multiple neuroimaging biomarkers is crucial for early detection of Alzheimer’s disease. The major challenges arise from the nonlinear relationship between biomarkers and assessment scores and the inter-correlation among them, which have not yet been well addressed. In this paper, we propose multi-layer multi-target regression (MMR) which enables simultaneously modeling intrinsic inter-target correlations and nonlinear input-output relationships in a general compositional framework. Specifically, by kernelized dictionary learning, the MMR can effectively handle highly nonlinear relationship between biomarkers and assessment scores; by robust low-rank linear learning via matrix elastic nets, the MMR can explicitly encode inter-correlations among multiple assessment scores; moreover, the MMR is flexibly and allows to work with non-smooth ℓ2,1-norm loss function, which enables calibration of multiple targets with disparate noise levels for more robust parameter estimation. The MMR can be efficiently solved by an alternating optimization algorithm via gradient descent with guaranteed convergence. The MMR has been evaluated by extensive experiments on the ADNI database with MRI data, and produced high accuracy surpassing previous regression models, which demonstrates its great effectiveness as a new multi-target regression model for clinical multivariate prediction.

KW - Alzheimer’s disease

KW - Calibration

KW - Multi-target regression

KW - Nonlinear regression

KW - Robust low-rank learning

UR - http://www.scopus.com/inward/record.url?scp=85047435460&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85047435460&partnerID=8YFLogxK

U2 - 10.1007/s12021-018-9381-1

DO - 10.1007/s12021-018-9381-1

M3 - Article

C2 - 29802511

AN - SCOPUS:85047435460

SP - 1

EP - 10

JO - Neuroinformatics

JF - Neuroinformatics

SN - 1539-2791

ER -