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

•A deep learning model, the deep canonically correlated sparse autoencoder model, is proposed for schizophrenia classification.•The deep learning model, with complex nonlinear transformation and sparsity, outperforms classical models.•Important and abstract features are extracted.•Classification res...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Computer methods and programs in biomedicine Jg. 183; S. 105073
Hauptverfasser: Li, Gang, Han, Depeng, Wang, Chao, Hu, Wenxing, Calhoun, Vince D., Wang, Yu-Ping
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier B.V 01.01.2020
Schlagworte:
ISSN:0169-2607, 1872-7565, 1872-7565
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•A deep learning model, the deep canonically correlated sparse autoencoder model, is proposed for schizophrenia classification.•The deep learning model, with complex nonlinear transformation and sparsity, outperforms classical models.•Important and abstract features are extracted.•Classification results are presented on single nucleotide polymorphisms (SNP) dataset and functional magnetic resonance imaging (fMRI) dataset from the mind clinical imaging consortium. 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. 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. 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. Experiments demonstrated higher accuracy of using the proposed method over other conventional models when classifying schizophrenia patients and healthy controls.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0169-2607
1872-7565
1872-7565
DOI:10.1016/j.cmpb.2019.105073