Graph auto-encoding brain networks with applications to analyzing large-scale brain imaging datasets

There has been a huge interest in studying human brain connectomes inferred from different imaging modalities and exploring their relationships with human traits, such as cognition. Brain connectomes are usually represented as networks, with nodes corresponding to different regions of interest (ROIs...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Meimei Liu, Zhengwu Zhang, David B. Dunson
Formato:	article
Lenguaje:	EN
Publicado:	Elsevier 2021
Materias:	Brain networks Non-linear factor analysis Graph CNN Replicated networks Variational auto-encoder Neurosciences. Biological psychiatry. Neuropsychiatry RC321-571
Acceso en línea:	https://doaj.org/article/64bfe1c9fbfd4bec91bbcc862e0d3f00
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:64bfe1c9fbfd4bec91bbcc862e0d3f00
record_format	dspace
spelling	oai:doaj.org-article:64bfe1c9fbfd4bec91bbcc862e0d3f002021-12-02T04:59:32ZGraph auto-encoding brain networks with applications to analyzing large-scale brain imaging datasets1095-957210.1016/j.neuroimage.2021.118750https://doaj.org/article/64bfe1c9fbfd4bec91bbcc862e0d3f002021-12-01T00:00:00Zhttp://www.sciencedirect.com/science/article/pii/S1053811921010223https://doaj.org/toc/1095-9572There has been a huge interest in studying human brain connectomes inferred from different imaging modalities and exploring their relationships with human traits, such as cognition. Brain connectomes are usually represented as networks, with nodes corresponding to different regions of interest (ROIs) and edges to connection strengths between ROIs. Due to the high-dimensionality and non-Euclidean nature of networks, it is challenging to depict their population distribution and relate them to human traits. Current approaches focus on summarizing the network using either pre-specified topological features or principal components analysis (PCA). In this paper, building on recent advances in deep learning, we develop a nonlinear latent factor model to characterize the population distribution of brain graphs and infer their relationships to human traits. We refer to our method as Graph AuTo-Encoding (GATE). We applied GATE to two large-scale brain imaging datasets, the Adolescent Brain Cognitive Development (ABCD) study and the Human Connectome Project (HCP) for adults, to study the structural brain connectome and its relationship with cognition. Numerical results demonstrate huge advantages of GATE over competitors in terms of prediction accuracy, statistical inference, and computing efficiency. We found that the structural connectome has a stronger association with a wide range of human cognitive traits than was apparent using previous approaches.Meimei LiuZhengwu ZhangDavid B. DunsonElsevierarticleBrain networksNon-linear factor analysisGraph CNNReplicated networksVariational auto-encoderNeurosciences. Biological psychiatry. NeuropsychiatryRC321-571ENNeuroImage, Vol 245, Iss , Pp 118750- (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Brain networks Non-linear factor analysis Graph CNN Replicated networks Variational auto-encoder Neurosciences. Biological psychiatry. Neuropsychiatry RC321-571
spellingShingle	Brain networks Non-linear factor analysis Graph CNN Replicated networks Variational auto-encoder Neurosciences. Biological psychiatry. Neuropsychiatry RC321-571 Meimei Liu Zhengwu Zhang David B. Dunson Graph auto-encoding brain networks with applications to analyzing large-scale brain imaging datasets
description	There has been a huge interest in studying human brain connectomes inferred from different imaging modalities and exploring their relationships with human traits, such as cognition. Brain connectomes are usually represented as networks, with nodes corresponding to different regions of interest (ROIs) and edges to connection strengths between ROIs. Due to the high-dimensionality and non-Euclidean nature of networks, it is challenging to depict their population distribution and relate them to human traits. Current approaches focus on summarizing the network using either pre-specified topological features or principal components analysis (PCA). In this paper, building on recent advances in deep learning, we develop a nonlinear latent factor model to characterize the population distribution of brain graphs and infer their relationships to human traits. We refer to our method as Graph AuTo-Encoding (GATE). We applied GATE to two large-scale brain imaging datasets, the Adolescent Brain Cognitive Development (ABCD) study and the Human Connectome Project (HCP) for adults, to study the structural brain connectome and its relationship with cognition. Numerical results demonstrate huge advantages of GATE over competitors in terms of prediction accuracy, statistical inference, and computing efficiency. We found that the structural connectome has a stronger association with a wide range of human cognitive traits than was apparent using previous approaches.
format	article
author	Meimei Liu Zhengwu Zhang David B. Dunson
author_facet	Meimei Liu Zhengwu Zhang David B. Dunson
author_sort	Meimei Liu
title	Graph auto-encoding brain networks with applications to analyzing large-scale brain imaging datasets
title_short	Graph auto-encoding brain networks with applications to analyzing large-scale brain imaging datasets
title_full	Graph auto-encoding brain networks with applications to analyzing large-scale brain imaging datasets
title_fullStr	Graph auto-encoding brain networks with applications to analyzing large-scale brain imaging datasets
title_full_unstemmed	Graph auto-encoding brain networks with applications to analyzing large-scale brain imaging datasets
title_sort	graph auto-encoding brain networks with applications to analyzing large-scale brain imaging datasets
publisher	Elsevier
publishDate	2021
url	https://doaj.org/article/64bfe1c9fbfd4bec91bbcc862e0d3f00
work_keys_str_mv	AT meimeiliu graphautoencodingbrainnetworkswithapplicationstoanalyzinglargescalebrainimagingdatasets AT zhengwuzhang graphautoencodingbrainnetworkswithapplicationstoanalyzinglargescalebrainimagingdatasets AT davidbdunson graphautoencodingbrainnetworkswithapplicationstoanalyzinglargescalebrainimagingdatasets
_version_	1718400892849356800

Graph auto-encoding brain networks with applications to analyzing large-scale brain imaging datasets

Ejemplares similares