High classification accuracy for schizophrenia with rest and task fMRI data

Wei Du, Vince Daniel Calhoun, Hualiang Li, Sai Ma, Tom Eichele, Kent A. Kiehl, Godfrey D. Pearlson, Tülay Adali

Research output: Contribution to journalArticle

Abstract

We present a novel method to extract classification features from functional magnetic resonance imaging (fMRI) data collected at rest or during the performance of a task. By combining a two-level feature identification scheme with kernel principal component analysis (KPCA) and Fisher's linear discriminant analysis (FLD), we achieve high classification rates in discriminating healthy controls from patients with schizophrenia. Experimental results using leave-one-out cross-validation show that features extracted from the default mode network (DMN) lead to a classification accuracy of over 90% in both data sets. Moreover, using a majority vote method that uses multiple features, we achieve a classification accuracy of 98% in auditory oddball (AOD) task and 93% in rest data. Several components, including DMN, temporal, and medial visual regions, are consistently present in the set of features that yield high classification accuracy. The features we have extracted thus show promise to be used as biomarkers for schizophrenia. Results also suggest that there may be different advantages to using resting fMRI data or task fMRI data.

Original languageEnglish (US)
Article number145
JournalFrontiers in Human Neuroscience
Issue numberJUNE 2012
DOIs
StatePublished - Jun 4 2012
Externally publishedYes

Fingerprint

Schizophrenia
Magnetic Resonance Imaging
Task Performance and Analysis
Discriminant Analysis
Principal Component Analysis
Biomarkers

Keywords

  • Classification
  • FLD
  • fMRI
  • Independent component analysis
  • KPCA

ASJC Scopus subject areas

  • Psychiatry and Mental health
  • Neurology
  • Biological Psychiatry
  • Behavioral Neuroscience
  • Neuropsychology and Physiological Psychology

Cite this

High classification accuracy for schizophrenia with rest and task fMRI data. / Du, Wei; Calhoun, Vince Daniel; Li, Hualiang; Ma, Sai; Eichele, Tom; Kiehl, Kent A.; Pearlson, Godfrey D.; Adali, Tülay.

In: Frontiers in Human Neuroscience, No. JUNE 2012, 145, 04.06.2012.

Research output: Contribution to journalArticle

Du, Wei ; Calhoun, Vince Daniel ; Li, Hualiang ; Ma, Sai ; Eichele, Tom ; Kiehl, Kent A. ; Pearlson, Godfrey D. ; Adali, Tülay. / High classification accuracy for schizophrenia with rest and task fMRI data. In: Frontiers in Human Neuroscience. 2012 ; No. JUNE 2012.
@article{527f86970283477e997acc01dc0e403f,
title = "High classification accuracy for schizophrenia with rest and task fMRI data",
abstract = "We present a novel method to extract classification features from functional magnetic resonance imaging (fMRI) data collected at rest or during the performance of a task. By combining a two-level feature identification scheme with kernel principal component analysis (KPCA) and Fisher's linear discriminant analysis (FLD), we achieve high classification rates in discriminating healthy controls from patients with schizophrenia. Experimental results using leave-one-out cross-validation show that features extracted from the default mode network (DMN) lead to a classification accuracy of over 90{\%} in both data sets. Moreover, using a majority vote method that uses multiple features, we achieve a classification accuracy of 98{\%} in auditory oddball (AOD) task and 93{\%} in rest data. Several components, including DMN, temporal, and medial visual regions, are consistently present in the set of features that yield high classification accuracy. The features we have extracted thus show promise to be used as biomarkers for schizophrenia. Results also suggest that there may be different advantages to using resting fMRI data or task fMRI data.",
keywords = "Classification, FLD, fMRI, Independent component analysis, KPCA",
author = "Wei Du and Calhoun, {Vince Daniel} and Hualiang Li and Sai Ma and Tom Eichele and Kiehl, {Kent A.} and Pearlson, {Godfrey D.} and T{\"u}lay Adali",
year = "2012",
month = "6",
day = "4",
doi = "10.3389/fnhum.2012.00145",
language = "English (US)",
journal = "Frontiers in Human Neuroscience",
issn = "1662-5161",
publisher = "Frontiers Research Foundation",
number = "JUNE 2012",

}

TY - JOUR

T1 - High classification accuracy for schizophrenia with rest and task fMRI data

AU - Du, Wei

AU - Calhoun, Vince Daniel

AU - Li, Hualiang

AU - Ma, Sai

AU - Eichele, Tom

AU - Kiehl, Kent A.

AU - Pearlson, Godfrey D.

AU - Adali, Tülay

PY - 2012/6/4

Y1 - 2012/6/4

N2 - We present a novel method to extract classification features from functional magnetic resonance imaging (fMRI) data collected at rest or during the performance of a task. By combining a two-level feature identification scheme with kernel principal component analysis (KPCA) and Fisher's linear discriminant analysis (FLD), we achieve high classification rates in discriminating healthy controls from patients with schizophrenia. Experimental results using leave-one-out cross-validation show that features extracted from the default mode network (DMN) lead to a classification accuracy of over 90% in both data sets. Moreover, using a majority vote method that uses multiple features, we achieve a classification accuracy of 98% in auditory oddball (AOD) task and 93% in rest data. Several components, including DMN, temporal, and medial visual regions, are consistently present in the set of features that yield high classification accuracy. The features we have extracted thus show promise to be used as biomarkers for schizophrenia. Results also suggest that there may be different advantages to using resting fMRI data or task fMRI data.

AB - We present a novel method to extract classification features from functional magnetic resonance imaging (fMRI) data collected at rest or during the performance of a task. By combining a two-level feature identification scheme with kernel principal component analysis (KPCA) and Fisher's linear discriminant analysis (FLD), we achieve high classification rates in discriminating healthy controls from patients with schizophrenia. Experimental results using leave-one-out cross-validation show that features extracted from the default mode network (DMN) lead to a classification accuracy of over 90% in both data sets. Moreover, using a majority vote method that uses multiple features, we achieve a classification accuracy of 98% in auditory oddball (AOD) task and 93% in rest data. Several components, including DMN, temporal, and medial visual regions, are consistently present in the set of features that yield high classification accuracy. The features we have extracted thus show promise to be used as biomarkers for schizophrenia. Results also suggest that there may be different advantages to using resting fMRI data or task fMRI data.

KW - Classification

KW - FLD

KW - fMRI

KW - Independent component analysis

KW - KPCA

UR - http://www.scopus.com/inward/record.url?scp=84864224120&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84864224120&partnerID=8YFLogxK

U2 - 10.3389/fnhum.2012.00145

DO - 10.3389/fnhum.2012.00145

M3 - Article

C2 - 22675292

AN - SCOPUS:84864224120

JO - Frontiers in Human Neuroscience

JF - Frontiers in Human Neuroscience

SN - 1662-5161

IS - JUNE 2012

M1 - 145

ER -