Application of deep canonically correlated sparse autoencoder for the classification of schizophrenia

Gang Li, Depeng Han, Chao Wang, Wenxing Hu, Vince Daniel Calhoun, Yu Ping Wang

Research output: Contribution to journalArticle

Abstract

Background and objective: Imaging genetics has been widely used to help diagnose and treat mental illness, e.g., schizophrenia, by combining magnetic resonance imaging of the brain and genomic information for comprehensive and systematic analysis. As a result, utilizing the correlation between magnetic resonance imaging of the brain and genomic information is becoming an important challenge. Methods: In this paper, the joint analysis of single nucleotide polymorphisms and functional magnetic resonance imaging is conducted for comprehensive study of schizophrenia. We developed a deep canonically correlated sparse autoencoder to classify schizophrenia patients from healthy controls, which can address the limitation of many existing methods such as canonical correlation analysis, deep canonical correlation analysis and sparse autoencoder. Results: The proposed deep canonically correlated sparse autoencoder can not only use complex nonlinear transformation and dimension reduction, but also achieve more accurate classifications. Our experiments showed the proposed method achieved an accuracy of 95.65% for SNP data sets and an accuracy of 80.53% for fMRI data sets. Conclusions: Experiments demonstrated higher accuracy of using the proposed method over other conventional models when classifying schizophrenia patients and healthy controls.

Original languageEnglish (US)
Article number105073
JournalComputer Methods and Programs in Biomedicine
Volume183
DOIs
StatePublished - Jan 1 2020

Fingerprint

Schizophrenia
Magnetic Resonance Imaging
Magnetic resonance
Imaging techniques
Brain
Single Nucleotide Polymorphism
Nucleotides
Polymorphism
Experiments
Datasets
Genetics

Keywords

  • Canonical correlation analysis
  • Deep canonically correlated sparse autoencoder
  • Imaging-genetic associations
  • Schizophrenia classification
  • Sparse autoencoder

ASJC Scopus subject areas

  • Software
  • Computer Science Applications
  • Health Informatics

Cite this

Application of deep canonically correlated sparse autoencoder for the classification of schizophrenia. / Li, Gang; Han, Depeng; Wang, Chao; Hu, Wenxing; Calhoun, Vince Daniel; Wang, Yu Ping.

In: Computer Methods and Programs in Biomedicine, Vol. 183, 105073, 01.01.2020.

Research output: Contribution to journalArticle

@article{0d40b4c1beda49169768a37f09a647a1,
title = "Application of deep canonically correlated sparse autoencoder for the classification of schizophrenia",
abstract = "Background and objective: Imaging genetics has been widely used to help diagnose and treat mental illness, e.g., schizophrenia, by combining magnetic resonance imaging of the brain and genomic information for comprehensive and systematic analysis. As a result, utilizing the correlation between magnetic resonance imaging of the brain and genomic information is becoming an important challenge. Methods: In this paper, the joint analysis of single nucleotide polymorphisms and functional magnetic resonance imaging is conducted for comprehensive study of schizophrenia. We developed a deep canonically correlated sparse autoencoder to classify schizophrenia patients from healthy controls, which can address the limitation of many existing methods such as canonical correlation analysis, deep canonical correlation analysis and sparse autoencoder. Results: The proposed deep canonically correlated sparse autoencoder can not only use complex nonlinear transformation and dimension reduction, but also achieve more accurate classifications. Our experiments showed the proposed method achieved an accuracy of 95.65{\%} for SNP data sets and an accuracy of 80.53{\%} for fMRI data sets. Conclusions: Experiments demonstrated higher accuracy of using the proposed method over other conventional models when classifying schizophrenia patients and healthy controls.",
keywords = "Canonical correlation analysis, Deep canonically correlated sparse autoencoder, Imaging-genetic associations, Schizophrenia classification, Sparse autoencoder",
author = "Gang Li and Depeng Han and Chao Wang and Wenxing Hu and Calhoun, {Vince Daniel} and Wang, {Yu Ping}",
year = "2020",
month = "1",
day = "1",
doi = "10.1016/j.cmpb.2019.105073",
language = "English (US)",
volume = "183",
journal = "Computer Methods and Programs in Biomedicine",
issn = "0169-2607",
publisher = "Elsevier Ireland Ltd",

}

TY - JOUR

T1 - Application of deep canonically correlated sparse autoencoder for the classification of schizophrenia

AU - Li, Gang

AU - Han, Depeng

AU - Wang, Chao

AU - Hu, Wenxing

AU - Calhoun, Vince Daniel

AU - Wang, Yu Ping

PY - 2020/1/1

Y1 - 2020/1/1

N2 - Background and objective: Imaging genetics has been widely used to help diagnose and treat mental illness, e.g., schizophrenia, by combining magnetic resonance imaging of the brain and genomic information for comprehensive and systematic analysis. As a result, utilizing the correlation between magnetic resonance imaging of the brain and genomic information is becoming an important challenge. Methods: In this paper, the joint analysis of single nucleotide polymorphisms and functional magnetic resonance imaging is conducted for comprehensive study of schizophrenia. We developed a deep canonically correlated sparse autoencoder to classify schizophrenia patients from healthy controls, which can address the limitation of many existing methods such as canonical correlation analysis, deep canonical correlation analysis and sparse autoencoder. Results: The proposed deep canonically correlated sparse autoencoder can not only use complex nonlinear transformation and dimension reduction, but also achieve more accurate classifications. Our experiments showed the proposed method achieved an accuracy of 95.65% for SNP data sets and an accuracy of 80.53% for fMRI data sets. Conclusions: Experiments demonstrated higher accuracy of using the proposed method over other conventional models when classifying schizophrenia patients and healthy controls.

AB - Background and objective: Imaging genetics has been widely used to help diagnose and treat mental illness, e.g., schizophrenia, by combining magnetic resonance imaging of the brain and genomic information for comprehensive and systematic analysis. As a result, utilizing the correlation between magnetic resonance imaging of the brain and genomic information is becoming an important challenge. Methods: In this paper, the joint analysis of single nucleotide polymorphisms and functional magnetic resonance imaging is conducted for comprehensive study of schizophrenia. We developed a deep canonically correlated sparse autoencoder to classify schizophrenia patients from healthy controls, which can address the limitation of many existing methods such as canonical correlation analysis, deep canonical correlation analysis and sparse autoencoder. Results: The proposed deep canonically correlated sparse autoencoder can not only use complex nonlinear transformation and dimension reduction, but also achieve more accurate classifications. Our experiments showed the proposed method achieved an accuracy of 95.65% for SNP data sets and an accuracy of 80.53% for fMRI data sets. Conclusions: Experiments demonstrated higher accuracy of using the proposed method over other conventional models when classifying schizophrenia patients and healthy controls.

KW - Canonical correlation analysis

KW - Deep canonically correlated sparse autoencoder

KW - Imaging-genetic associations

KW - Schizophrenia classification

KW - Sparse autoencoder

UR - http://www.scopus.com/inward/record.url?scp=85072091523&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85072091523&partnerID=8YFLogxK

U2 - 10.1016/j.cmpb.2019.105073

DO - 10.1016/j.cmpb.2019.105073

M3 - Article

VL - 183

JO - Computer Methods and Programs in Biomedicine

JF - Computer Methods and Programs in Biomedicine

SN - 0169-2607

M1 - 105073

ER -