Simultaneous Differential Network Analysis and Classification for High-dimensional Matrix-variate Data, with application to Brain Connectivity Alteration Detection and fMRI-guided Medical Diagnoses of Alzheimer's Disease

TL;DR

This paper introduces a novel ensemble-learning framework for differential network analysis and classification of high-dimensional matrix-variate fMRI data, effectively identifying brain connectivity alterations associated with Alzheimer's disease.

Contribution

It develops a Kronecker product covariance model and an ensemble-learning method to simultaneously analyze differential brain networks and classify Alzheimer's disease from matrix-variate data.

Findings

Identified consistent hub nodes and differential interaction patterns in AD.

Achieved high out-of-sample classification accuracy.

Validated the method on real fMRI data with results aligning with existing studies.

Abstract

Alzheimer's disease (AD) is the most common form of dementia, which causes problems with memory, thinking and behavior. Growing evidence has shown that the brain connectivity network experiences alterations for such a complex disease. Network comparison, also known as differential network analysis, is thus particularly powerful to reveal the disease pathologies and identify clinical biomarkers for medical diagnoses (classification). Data from neurophysiological measurements are multi-dimensional and in matrix-form, which poses major challenges in brain connectivity analysis and medical diagnoses. Naive vectorization method is not sufficient as it ignores the structural information within the matrix. In the article, we adopt the Kronecker product covariance matrix framework to capture both spatial and temporal correlations of the matrix-variate data while the temporal covariance matrix is treated as a nuisance parameter. By recognizing that the strengths of network connections may vary across subjects, we develop an ensemble-learning procedure, which identifies the differential interaction patterns of brain regions between the AD group and the control group and conducts medical diagnosis (classification) of AD simultaneously. We applied the proposed procedure to functional connectivity analysis of fMRI dataset related with Alzheimer's disease. The hub nodes and differential interaction patterns identified are consistent with existing experimental studies, and satisfactory out-of-sample classification performance is achieved for medical diagnosis of Alzheimer's disease. An R package \SDNCMV" for implementation is available at https://github.com/heyongstat/SDNCMV.