Nonlinear ICA of fMRI reveals primitive temporal structures linked to rest, task, and behavioral traits

Show full item record



Permalink

http://hdl.handle.net/10138/321137

Citation

Morioka , H , Calhoun , V & Hyvarinen , A 2020 , ' Nonlinear ICA of fMRI reveals primitive temporal structures linked to rest, task, and behavioral traits ' , NeuroImage , vol. 218 , 116989 . https://doi.org/10.1016/j.neuroimage.2020.116989

Title: Nonlinear ICA of fMRI reveals primitive temporal structures linked to rest, task, and behavioral traits
Author: Morioka, Hiroshi; Calhoun, Vince; Hyvarinen, Aapo
Other contributor: University of Helsinki, Department of Computer Science


Date: 2020-09
Language: eng
Number of pages: 17
Belongs to series: NeuroImage
ISSN: 1053-8119
DOI: https://doi.org/10.1016/j.neuroimage.2020.116989
URI: http://hdl.handle.net/10138/321137
Abstract: Accumulating evidence from whole brain functional magnetic resonance imaging (fMRI) suggests that the human brain at rest is functionally organized in a spatially and temporally constrained manner. However, because of their complexity, the fundamental mechanisms underlying time-varying functional networks are still not well under-stood. Here, we develop a novel nonlinear feature extraction framework called local space-contrastive learning (LSCL), which extracts distinctive nonlinear temporal structure hidden in time series, by training a deep temporal convolutional neural network in an unsupervised, data-driven manner. We demonstrate that LSCL identifies certain distinctive local temporal structures, referred to as temporal primitives, which repeatedly appear at different time points and spatial locations, reflecting dynamic resting-state networks. We also show that these temporal primitives are also present in task-evoked spatiotemporal responses. We further show that the temporal primitives capture unique aspects of behavioral traits such as fluid intelligence and working memory. These re-sults highlight the importance of capturing transient spatiotemporal dynamics within fMRI data and suggest that such temporal primitives may capture fundamental information underlying both spontaneous and task-induced fMRI dynamics.
Subject: Nonlinear spatial independent component analysis (sICA)
Local space-contrastive learning (LSCL)
Unsupervised deep learning
Temporal primitives
Resting-state functional magnetic resonance imaging (fMRI)
Behavioral traits
INDEPENDENT COMPONENT ANALYSIS
DYNAMIC FUNCTIONAL CONNECTIVITY
BRAIN CONNECTIVITY
STATE FMRI
FLUCTUATIONS
STIMULUS
ORGANIZATION
RESPIRATION
NETWORKS
PATTERNS
113 Computer and information sciences
Rights:


Files in this item

Total number of downloads: Loading...

Files Size Format View
1_s2.0_S1053811920304754_main.pdf 6.328Mb PDF View/Open

This item appears in the following Collection(s)

Show full item record