Major depressive disorder identification by referenced multiset canonical correlation analysis with clinical scores

• Published in: Medical image analysis Vol. 60; p. 101600
• Main Authors: Lin, Wuhong, Lv, Dongsheng, Han, Ziliang, Dong, Jianwei, Yang, Lihua
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.02.2020; Elsevier BV
• Subjects: Canonical correlation analysis; Classification; Connectome - methods; Correlation analysis; Depressive Disorder, Major - diagnostic imaging; Discriminant Analysis; Feature extraction; Functional connection; Functional magnetic resonance imaging; Humans; Linear discrimination analysis; Magnetic Resonance Imaging; Major depressive disorder identification; Mental depression; Model accuracy; Multivariate Analysis; Regularization; Signal to noise ratio; Functional magnetic resonance imaging; Functional connection; Linear discrimination analysis; Canonical correlation analysis; Major depressive disorder identification
• ISSN: 1361-8415; 1361-8423
• DOI: 10.1016/j.media.2019.101600

•The frequency of the brain activity is used to compute functional connections.•The functional connections are described in several modes.•Clinical information is fused in selecting classification features.•Only a small subset of connections are selected for classification. [Display omitted] A novel method based on multiset canonical correlation analysis (mCCA) and linear discriminant analysis (LDA) is presented to identify the major depressive disorder (MDD). The new method comprises two parts, namely, the mCCA-rreg and sparse LDA models. The mCCA-rreg model extends the classical canonical correlation model to calculate functional connections by restricting the references to a reference space and adding a spatial regularization term. The reference space is used to ensure that the model extracts important components first from several datasets simultaneously by decreasing the importance of the components in which we are uninterested. The spatial regularization term helps in avoiding the multicollinearity and overfitting problems under the low signal-to-noise ratio circumstance. The sparse LDA model extends the classical LDA model to extract a small subset of discriminative classification features by fusing clinical scores. In the real data experiment, we extract two functional connection modes from 45 subjects by the mCCA-rreg model. Then, we construct classifiers to identify the patients with MDD based on the connections selected by the sparse LDA model. The best accuracy is higher than 95%. The results show that the mCCA-rreg model can retrieve the important components characterized by a preassigned reference space and exclude the noise or components of no interest. The sparse LDA model can extract discriminative classification features related to clinical scores.