Multi-site clustering and nested feature extraction for identifying autism spectrum disorder with resting-state fMRI

• Published in: Medical image analysis Vol. 75; p. 102279
• Main Authors: Wang, Nan, Yao, Dongren, Ma, Lizhuang, Liu, Mingxia
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.01.2022; Elsevier BV
• Subjects: Autism; Autism spectrum disorder; Autism Spectrum Disorder - diagnostic imaging; Biomarkers; Brain - diagnostic imaging; Brain Mapping; Cerebellum; Cluster Analysis; Clustering; Discriminative biomarker identification; Feature extraction; fMRI; Functional connectivity; Functional magnetic resonance imaging; Heterogeneity; Humans; Magnetic Resonance Imaging; Medical imaging; Mental disorders; Neural networks; Neuroimaging; Population studies; Singular value decomposition; Support vector machines; Training; Functional connectivity; fMRI; Discriminative biomarker identification; Autism spectrum disorder
• ISSN: 1361-8415; 1361-8423; 1361-8423
• DOI: 10.1016/j.media.2021.102279

•We propose a general framework to model inter-site heterogeneity for functional connectivity based brain disease identification, and it can be directly applied to other multi-site applications with resting-state functional MRI (rs-fMRI) data.•We design a nested singular value decomposition (SVD) method to mitigate inter-site data heterogeneity and extract functional connectivity features, by learning both local cluster-shared features across sites for each group/category and global category-shared features within a specifici group (i.e., the ASD patient or healthy control group).•The proposed method helps to identify disease-associated functional connectivity abnormality from rs-fMRI data, so it has good interpretability. [Display omitted] Brain functional connectivity (FC) derived from resting-state functional magnetic resonance imaging (rs-fMRI) has been widely employed to study neuropsychiatric disorders such as autism spectrum disorder (ASD). Existing studies usually suffer from (1) significant data heterogeneity caused by different scanners or studied populations in multiple sites, (2) curse of dimensionality caused by millions of voxels in each fMRI scan and a very limited number (tens or hundreds) of training samples, and (3) poor interpretability, which hinders the identification of reproducible disease biomarkers. To this end, we propose a Multi-site Clustering and Nested Feature Extraction (MC-NFE) method for fMRI-based ASD detection. Specifically, we first divide multi-site training data into ASD and healthy control (HC) groups. To model inter-site heterogeneity within each category, we use a similarity-driven multiview linear reconstruction model to learn latent representations and perform subject clustering within each group. We then design a nested singular value decomposition (SVD) method to mitigate inter-site heterogeneity and extract FC features by learning both local cluster-shared features across sites within each category and global category-shared features across ASD and HC groups, followed by a linear support vector machine (SVM) for ASD detection. Experimental results on 609 subjects with rs-fMRI from the ABIDE database with 21 imaging sites suggest that the proposed MC-NFE outperforms several state-of-the-art methods in ASD detection. The most discriminative FCs identified by the MC-NFE are mainly located in default mode network, salience network, and cerebellum region, which could be used as potential biomarkers for fMRI-based ASD analysis.