Brain network decomposition by Auto Encoder (AE) and graph auto encoder (GAE)

Myungwon Choi, Pilsub Lee, Daegyeom Kim, Suji Lee, Hyun Chul Youn, Hyun Ghang Jeong, Cheol E. Han

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Brain networks consist of nodes that are anatomically defined brain regions, and edges that connect a pair of brain regions. The diffusion-weighted magnetic resonance images and the advances in computer-aided tractography algorithms showed that human brain networks are strongly associated with cognitive functions. Brain regions dedicated to a specific cognitive function are spatially clustered and efficiently connected each other; this is called local functional segregation. However, it is not well known that such a local segregation is associated with sub-networks which may act as building blocks of brain networks. In this work, we used machine learning techniques to analyze brain networks. Specifically, using an auto-encoder and a graph auto-encoder, we decomposed brain networks into several essential building blocks, and compared their results through various measures of decomposition quality. We observed that the graph auto-encoder out-performed the auto-encoder, and that its results showed significant correlation with cognitive deterioration in Alzheimer’s disease.

Original languageEnglish
Title of host publicationNeural Information Processing - 26th International Conference, ICONIP 2019, Proceedings
EditorsTom Gedeon, Kok Wai Wong, Minho Lee
PublisherSpringer
Pages568-579
Number of pages12
ISBN (Print)9783030367077
DOIs
Publication statusPublished - 2019 Jan 1
Event26th International Conference on Neural Information Processing, ICONIP 2019 - Sydney, Australia
Duration: 2019 Dec 122019 Dec 15

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11953 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference26th International Conference on Neural Information Processing, ICONIP 2019
CountryAustralia
CitySydney
Period19/12/1219/12/15

Fingerprint

Encoder
Brain
Decomposition
Decompose
Graph in graph theory
Segregation
Building Blocks
Magnetic Resonance Image
Alzheimer's Disease
Magnetic resonance
Deterioration
Learning systems
Machine Learning
Vertex of a graph

Keywords

  • Alzheimer’s disease
  • Brain networks
  • Graph auto-encoder
  • Graph convolutional neural network

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Choi, M., Lee, P., Kim, D., Lee, S., Youn, H. C., Jeong, H. G., & Han, C. E. (2019). Brain network decomposition by Auto Encoder (AE) and graph auto encoder (GAE). In T. Gedeon, K. W. Wong, & M. Lee (Eds.), Neural Information Processing - 26th International Conference, ICONIP 2019, Proceedings (pp. 568-579). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11953 LNCS). Springer. https://doi.org/10.1007/978-3-030-36708-4_47

Brain network decomposition by Auto Encoder (AE) and graph auto encoder (GAE). / Choi, Myungwon; Lee, Pilsub; Kim, Daegyeom; Lee, Suji; Youn, Hyun Chul; Jeong, Hyun Ghang; Han, Cheol E.

Neural Information Processing - 26th International Conference, ICONIP 2019, Proceedings. ed. / Tom Gedeon; Kok Wai Wong; Minho Lee. Springer, 2019. p. 568-579 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11953 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Choi, M, Lee, P, Kim, D, Lee, S, Youn, HC, Jeong, HG & Han, CE 2019, Brain network decomposition by Auto Encoder (AE) and graph auto encoder (GAE). in T Gedeon, KW Wong & M Lee (eds), Neural Information Processing - 26th International Conference, ICONIP 2019, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11953 LNCS, Springer, pp. 568-579, 26th International Conference on Neural Information Processing, ICONIP 2019, Sydney, Australia, 19/12/12. https://doi.org/10.1007/978-3-030-36708-4_47
Choi M, Lee P, Kim D, Lee S, Youn HC, Jeong HG et al. Brain network decomposition by Auto Encoder (AE) and graph auto encoder (GAE). In Gedeon T, Wong KW, Lee M, editors, Neural Information Processing - 26th International Conference, ICONIP 2019, Proceedings. Springer. 2019. p. 568-579. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-030-36708-4_47
Choi, Myungwon ; Lee, Pilsub ; Kim, Daegyeom ; Lee, Suji ; Youn, Hyun Chul ; Jeong, Hyun Ghang ; Han, Cheol E. / Brain network decomposition by Auto Encoder (AE) and graph auto encoder (GAE). Neural Information Processing - 26th International Conference, ICONIP 2019, Proceedings. editor / Tom Gedeon ; Kok Wai Wong ; Minho Lee. Springer, 2019. pp. 568-579 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{03a6957346534fe88be987bbd8b4a3d2,
title = "Brain network decomposition by Auto Encoder (AE) and graph auto encoder (GAE)",
abstract = "Brain networks consist of nodes that are anatomically defined brain regions, and edges that connect a pair of brain regions. The diffusion-weighted magnetic resonance images and the advances in computer-aided tractography algorithms showed that human brain networks are strongly associated with cognitive functions. Brain regions dedicated to a specific cognitive function are spatially clustered and efficiently connected each other; this is called local functional segregation. However, it is not well known that such a local segregation is associated with sub-networks which may act as building blocks of brain networks. In this work, we used machine learning techniques to analyze brain networks. Specifically, using an auto-encoder and a graph auto-encoder, we decomposed brain networks into several essential building blocks, and compared their results through various measures of decomposition quality. We observed that the graph auto-encoder out-performed the auto-encoder, and that its results showed significant correlation with cognitive deterioration in Alzheimer’s disease.",
keywords = "Alzheimer’s disease, Brain networks, Graph auto-encoder, Graph convolutional neural network",
author = "Myungwon Choi and Pilsub Lee and Daegyeom Kim and Suji Lee and Youn, {Hyun Chul} and Jeong, {Hyun Ghang} and Han, {Cheol E.}",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/978-3-030-36708-4_47",
language = "English",
isbn = "9783030367077",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer",
pages = "568--579",
editor = "Tom Gedeon and Wong, {Kok Wai} and Minho Lee",
booktitle = "Neural Information Processing - 26th International Conference, ICONIP 2019, Proceedings",

}

TY - GEN

T1 - Brain network decomposition by Auto Encoder (AE) and graph auto encoder (GAE)

AU - Choi, Myungwon

AU - Lee, Pilsub

AU - Kim, Daegyeom

AU - Lee, Suji

AU - Youn, Hyun Chul

AU - Jeong, Hyun Ghang

AU - Han, Cheol E.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Brain networks consist of nodes that are anatomically defined brain regions, and edges that connect a pair of brain regions. The diffusion-weighted magnetic resonance images and the advances in computer-aided tractography algorithms showed that human brain networks are strongly associated with cognitive functions. Brain regions dedicated to a specific cognitive function are spatially clustered and efficiently connected each other; this is called local functional segregation. However, it is not well known that such a local segregation is associated with sub-networks which may act as building blocks of brain networks. In this work, we used machine learning techniques to analyze brain networks. Specifically, using an auto-encoder and a graph auto-encoder, we decomposed brain networks into several essential building blocks, and compared their results through various measures of decomposition quality. We observed that the graph auto-encoder out-performed the auto-encoder, and that its results showed significant correlation with cognitive deterioration in Alzheimer’s disease.

AB - Brain networks consist of nodes that are anatomically defined brain regions, and edges that connect a pair of brain regions. The diffusion-weighted magnetic resonance images and the advances in computer-aided tractography algorithms showed that human brain networks are strongly associated with cognitive functions. Brain regions dedicated to a specific cognitive function are spatially clustered and efficiently connected each other; this is called local functional segregation. However, it is not well known that such a local segregation is associated with sub-networks which may act as building blocks of brain networks. In this work, we used machine learning techniques to analyze brain networks. Specifically, using an auto-encoder and a graph auto-encoder, we decomposed brain networks into several essential building blocks, and compared their results through various measures of decomposition quality. We observed that the graph auto-encoder out-performed the auto-encoder, and that its results showed significant correlation with cognitive deterioration in Alzheimer’s disease.

KW - Alzheimer’s disease

KW - Brain networks

KW - Graph auto-encoder

KW - Graph convolutional neural network

UR - http://www.scopus.com/inward/record.url?scp=85077507361&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85077507361&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-36708-4_47

DO - 10.1007/978-3-030-36708-4_47

M3 - Conference contribution

AN - SCOPUS:85077507361

SN - 9783030367077

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 568

EP - 579

BT - Neural Information Processing - 26th International Conference, ICONIP 2019, Proceedings

A2 - Gedeon, Tom

A2 - Wong, Kok Wai

A2 - Lee, Minho

PB - Springer

ER -