A robust sparse-modeling framework for estimating schizophrenia biomarkers from fMRI

Keith Dillon, Vince Calhoun, Yu Ping Wang

Abstract

Background Our goal is to identify the brain regions most relevant to mental illness using neuroimaging. State of the art machine learning methods commonly suffer from repeatability difficulties in this application, particularly when using large and heterogeneous populations for samples. New method We revisit both dimensionality reduction and sparse modeling, and recast them in a common optimization-based framework. This allows us to combine the benefits of both types of methods in an approach which we call unambiguous components. We use this to estimate the image component with a constrained variability, which is best correlated with the unknown disease mechanism. Results We apply the method to the estimation of neuroimaging biomarkers for schizophrenia, using task fMRI data from a large multi-site study. The proposed approach yields an improvement in both robustness of the estimate and classification accuracy. Comparison with existing methods We find that unambiguous components incorporate roughly two thirds of the same brain regions as sparsity-based methods LASSO and elastic net, while roughly one third of the selected regions differ. Further, unambiguous components achieve superior classification accuracy in differentiating cases from controls. Conclusions Unambiguous components provide a robust way to estimate important regions of imaging data.

Original languageEnglish (US)
Pages (from-to)46-55
Number of pages10
JournalJournal of Neuroscience Methods
Volume276
DOIs
StatePublished - Jan 30 2017

Fingerprint

Schizophrenia
Biological Markers
Magnetic Resonance Imaging
Neuroimaging
Brain
Artificial Intelligence

Keywords

  • Functional MRI
  • Optimization
  • PCA
  • Schizophrenia
  • Sparsity

ASJC Scopus subject areas

  • Neuroscience(all)

Cite this

A robust sparse-modeling framework for estimating schizophrenia biomarkers from fMRI. / Dillon, Keith; Calhoun, Vince; Wang, Yu Ping.

In: Journal of Neuroscience Methods, Vol. 276, 30.01.2017, p. 46-55.

Research output: Contribution to journalArticle

Dillon, Keith; Calhoun, Vince; Wang, Yu Ping / A robust sparse-modeling framework for estimating schizophrenia biomarkers from fMRI.

In: Journal of Neuroscience Methods, Vol. 276, 30.01.2017, p. 46-55.

Research output: Contribution to journalArticle

@article{331c08c495bc4e4d98de30c4370f650b,
title = "A robust sparse-modeling framework for estimating schizophrenia biomarkers from fMRI",
keywords = "Functional MRI, Optimization, PCA, Schizophrenia, Sparsity",
author = "Keith Dillon and Vince Calhoun and Wang, {Yu Ping}",
year = "2017",
month = "1",
doi = "10.1016/j.jneumeth.2016.11.005",
volume = "276",
pages = "46--55",
journal = "Journal of Neuroscience Methods",
issn = "0165-0270",
publisher = "Elsevier",

}

TY - JOUR

T1 - A robust sparse-modeling framework for estimating schizophrenia biomarkers from fMRI

AU - Dillon,Keith

AU - Calhoun,Vince

AU - Wang,Yu Ping

PY - 2017/1/30

Y1 - 2017/1/30

N2 - Background Our goal is to identify the brain regions most relevant to mental illness using neuroimaging. State of the art machine learning methods commonly suffer from repeatability difficulties in this application, particularly when using large and heterogeneous populations for samples. New method We revisit both dimensionality reduction and sparse modeling, and recast them in a common optimization-based framework. This allows us to combine the benefits of both types of methods in an approach which we call unambiguous components. We use this to estimate the image component with a constrained variability, which is best correlated with the unknown disease mechanism. Results We apply the method to the estimation of neuroimaging biomarkers for schizophrenia, using task fMRI data from a large multi-site study. The proposed approach yields an improvement in both robustness of the estimate and classification accuracy. Comparison with existing methods We find that unambiguous components incorporate roughly two thirds of the same brain regions as sparsity-based methods LASSO and elastic net, while roughly one third of the selected regions differ. Further, unambiguous components achieve superior classification accuracy in differentiating cases from controls. Conclusions Unambiguous components provide a robust way to estimate important regions of imaging data.

AB - Background Our goal is to identify the brain regions most relevant to mental illness using neuroimaging. State of the art machine learning methods commonly suffer from repeatability difficulties in this application, particularly when using large and heterogeneous populations for samples. New method We revisit both dimensionality reduction and sparse modeling, and recast them in a common optimization-based framework. This allows us to combine the benefits of both types of methods in an approach which we call unambiguous components. We use this to estimate the image component with a constrained variability, which is best correlated with the unknown disease mechanism. Results We apply the method to the estimation of neuroimaging biomarkers for schizophrenia, using task fMRI data from a large multi-site study. The proposed approach yields an improvement in both robustness of the estimate and classification accuracy. Comparison with existing methods We find that unambiguous components incorporate roughly two thirds of the same brain regions as sparsity-based methods LASSO and elastic net, while roughly one third of the selected regions differ. Further, unambiguous components achieve superior classification accuracy in differentiating cases from controls. Conclusions Unambiguous components provide a robust way to estimate important regions of imaging data.

KW - Functional MRI

KW - Optimization

KW - PCA

KW - Schizophrenia

KW - Sparsity

UR - http://www.scopus.com/inward/record.url?scp=85000578211&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85000578211&partnerID=8YFLogxK

U2 - 10.1016/j.jneumeth.2016.11.005

DO - 10.1016/j.jneumeth.2016.11.005

M3 - Article

VL - 276

SP - 46

EP - 55

JO - Journal of Neuroscience Methods

T2 - Journal of Neuroscience Methods

JF - Journal of Neuroscience Methods

SN - 0165-0270

ER -