Classification of schizophrenia based on deep canonically correlated sparse autoencoder(PDF)
《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]
- Issue:
- 2020年第3期
- Page:
- 391-396
- Research Field:
- 其他（激光医学等）
- Publishing date:
Info
- Title:
- Classification of schizophrenia based on deep canonically correlated sparse autoencoder
- Author(s):
- LI Gang; HAN Depeng; LIU Qiangwei; WANG Chao; LI Ying
- School of Electronic Control, Chang’an University, Xi’an 710064, China
- Keywords:
- Keywords: schizophrenia; single nucleotide polymorphism; functional magnetic resonance imaging; deep canonically correlated sparse autoencoder
- PACS:
- R318；R749.3
- DOI:
- DOI:10.3969/j.issn.1005-202X.2020.03.025
- Abstract:
- Abstract: Imaging genetics has been widely used to aid the diagnosis and treatment of mental illnesses (e.g. schizophrenia), by combining magnetic resonance imaging of the brain and genomic information for a comprehensive and systematic analysis. Herein the joint analysis of single nucleotide polymorphism data and functional magnetic resonance imaging data is carried out for the comprehensive study on schizophrenia. Deep canonically correlated sparse autoencoder is proposed to explore the nonlinear relationships between two types of data and reduce dimensions, and then classify schizophrenia patients from healthy controls. The experimental results reveal that compared with other conventional models, deep canonically correlated sparse autoencoder model can achieve a higher classification accuracy.
Last Update: 2020-04-02