Diagnosis of Autism Spectrum Disorder Using Central-Moment Features From Low- and High-Order Dynamic Resting-State Functional Connectivity Networks.

Feng Zhao,Zhiyuan Chen,Islem Rekik,Seong-Whan Lee,Dinggang Shen

doi:10.3389/fnins.2020.00258

Abstract

The sliding-window-based dynamic functional connectivity networks (D-FCNs) derived from resting-state functional magnetic resonance imaging (rs-fMRI) are effective methods for diagnosing various neurological diseases, including autism spectrum disorder (ASD). However, traditional D-FCNs are low-order networks based on pairwise correlation between brain regions, thus overlooking high-level interactions across multiple regions of interest (ROIs). Moreover, D-FCNs suffer from the temporal mismatching issue, i.e., subnetworks in the same temporal window do not have temporal correspondence across different subjects. To address the above problems, we first construct a novel high-order D-FCNs based on the principle of “correlation’s correlation” to further explore the higher level and more complex interaction relationships among multiple ROIs. Furthermore, we propose to use a central-moment method to extract temporal-invariance properties contained in either low- or high-order D-FCNs. Finally, we design and train an ensemble classifier by fusing the features extracted from conventional FCN, low-order D-FCNs, and high-order D-FCNs for the diagnosis of ASD and normal control subjects. Our method achieved the best ASD classification accuracy (83%), and our results revealed the features extracted from different networks fingerprinting the autistic brain at different connectional levels.

Highlights

Autism spectrum disorder (ASD) is a serious childhood neurodevelopmental disease, characterized by the impairment in social interaction, communication, and many other behavioral and cognitive functions in varying degrees (Geschwind and Levitt, 2007)
There are three parts of contribution in this paper: (Geschwind and Levitt, 2007) proposing new Ho-dynamic functional connectivity networks (D-FC network (FCN)) to reflect high-level connectivity information across multiple regions of interest (ROIs); (Jin et al, 2015) utilizing a central-moment method to capture functional connectivity (FC) properties derived from Lo-D-FCNs or Ho-D-FCNs without performing chronological time matching; (Anagnostou and Taylor, 2011) employing three multilevel FCN models (i.e., conventional FCN (C-FCN), Lo-D-FCNs, and Ho-DFCNs) to comprehensively investigate complex and multilevel functional associations among brain ROIs
We select the combination of parameters that can lead to the highest ACC from the support vector machines (SVM) of C-FCN, Lo-D-FCNs, and HoD-FCNs, respectively, and obtain the final classification result by linear fusion of the SVM ensemble decision scores

Summary

INTRODUCTION

Autism spectrum disorder (ASD) is a serious childhood neurodevelopmental disease, characterized by the impairment in social interaction, communication, and many other behavioral and cognitive functions in varying degrees (Geschwind and Levitt, 2007). There are three parts of contribution in this paper: (Geschwind and Levitt, 2007) proposing new Ho-D-FCNs (never used in previous ASD diagnosis) to reflect high-level connectivity information across multiple ROIs; (Jin et al, 2015) utilizing a central-moment method to capture FC properties derived from Lo-D-FCNs or Ho-D-FCNs without performing chronological time matching; (Anagnostou and Taylor, 2011) employing three multilevel FCN models (i.e., C-FCN, Lo-D-FCNs, and Ho-DFCNs) to comprehensively investigate complex and multilevel functional associations among brain ROIs. The rs-fMRI dataset used in this paper was downloaded from a publicly available Autism Brain Imaging Data Exchange (ABIDE) database (Di Martino et al, 2013).

METHOD

DATA AVAILABILITY STATEMENT

CONCLUSION