BrainNetGAN: Data Augmentation of Brain Connectivity Using Generative Adversarial Network for Dementia Classification

Chao Li; Yiran Wei; Xi Chen; Carola Bibiane Schönlieb

doi:10.1007/978-3-030-88210-5_9

BrainNetGAN: Data Augmentation of Brain Connectivity Using Generative Adversarial Network for Dementia Classification

Chao Li, Yiran Wei, Xi Chen, Carola Bibiane Schönlieb

Research output: Chapter or section in a book/report/conference proceeding › Chapter in a published conference proceeding

7 Citations (SciVal)

97 Downloads (Pure)

Abstract

Alzheimer’s disease (AD) is the most common age-related dementia, which significantly affects an individual’s daily life and impact socioeconomics. It remains a challenge to identify the individuals at risk of dementia for precise management. Brain MRI offers a non-invasive biomarker to detect brain aging. Previous evidence shows that the structural brain network generated from the diffusion MRI promises to classify dementia accurately based on deep learning models. However, the limited availability of diffusion MRI challenges the model training of deep learning. We propose the BrainNetGAN, a variant of the generative adversarial network, to efficiently augment the structural brain networks for dementia classifying tasks. The BrainNetGAN model is trained to generate fake brain connectivity matrices, which are expected to reflect the latent distribution and topological features of the real brain network data. Numerical results show that the BrainNetGAN outperforms the benchmarking algorithms in augmenting the brain networks for AD classification tasks.

Original language	English
Title of host publication	Deep Generative Models, and Data Augmentation, Labelling, and Imperfections - First Workshop, DGM4MICCAI 2021, and First Workshop, DALI 2021, Held in Conjunction with MICCAI 2021, Proceedings
Editors	Sandy Engelhardt, Ilkay Oksuz, Dajiang Zhu, Yixuan Yuan, Anirban Mukhopadhyay, Nicholas Heller, Sharon Xiaolei Huang, Hien Nguyen, Raphael Sznitman, Yuan Xue
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	103-111
Number of pages	9
ISBN (Print)	9783030882099
DOIs	https://doi.org/10.1007/978-3-030-88210-5_9
Publication status	Published - 31 Dec 2021
Event	1st Workshop on Deep Generative Models for Medical Image Computing and Computer Assisted Intervention, DGM4MICCAI 2021 and 1st Workshop on Data Augmentation, Labelling, and Imperfections, DALI 2021 held in conjunction with 24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021 - Virtual, Online Duration: 1 Oct 2021 → 1 Oct 2021

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	13003 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	1st Workshop on Deep Generative Models for Medical Image Computing and Computer Assisted Intervention, DGM4MICCAI 2021 and 1st Workshop on Data Augmentation, Labelling, and Imperfections, DALI 2021 held in conjunction with 24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021
City	Virtual, Online
Period	1/10/21 → 1/10/21

Keywords

Brain connectivity
Classification
Data augmentation
Generative adversarial network

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-030-88210-5_9

2103.08494.PDF
This is a post-peer-review, pre-copyedit version of an article published in Lecture Notes in Computer Science. The final authenticated version is available online at: https://doi.org/10.1007/978-3-030-88210-5_9
Accepted author manuscript, 2.71 MB

https://arxiv.org/abs/2103.08494Licence: CC BY-NC-ND

Cite this

Li, C., Wei, Y., Chen, X., & Schönlieb, C. B. (2021). BrainNetGAN: Data Augmentation of Brain Connectivity Using Generative Adversarial Network for Dementia Classification. In S. Engelhardt, I. Oksuz, D. Zhu, Y. Yuan, A. Mukhopadhyay, N. Heller, S. X. Huang, H. Nguyen, R. Sznitman, & Y. Xue (Eds.), Deep Generative Models, and Data Augmentation, Labelling, and Imperfections - First Workshop, DGM4MICCAI 2021, and First Workshop, DALI 2021, Held in Conjunction with MICCAI 2021, Proceedings (pp. 103-111). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13003 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-88210-5_9

BrainNetGAN: Data Augmentation of Brain Connectivity Using Generative Adversarial Network for Dementia Classification. / Li, Chao; Wei, Yiran; Chen, Xi et al.
Deep Generative Models, and Data Augmentation, Labelling, and Imperfections - First Workshop, DGM4MICCAI 2021, and First Workshop, DALI 2021, Held in Conjunction with MICCAI 2021, Proceedings. ed. / Sandy Engelhardt; Ilkay Oksuz; Dajiang Zhu; Yixuan Yuan; Anirban Mukhopadhyay; Nicholas Heller; Sharon Xiaolei Huang; Hien Nguyen; Raphael Sznitman; Yuan Xue. Springer Science and Business Media Deutschland GmbH, 2021. p. 103-111 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13003 LNCS).

Research output: Chapter or section in a book/report/conference proceeding › Chapter in a published conference proceeding

Li, C, Wei, Y, Chen, X & Schönlieb, CB 2021, BrainNetGAN: Data Augmentation of Brain Connectivity Using Generative Adversarial Network for Dementia Classification. in S Engelhardt, I Oksuz, D Zhu, Y Yuan, A Mukhopadhyay, N Heller, SX Huang, H Nguyen, R Sznitman & Y Xue (eds), Deep Generative Models, and Data Augmentation, Labelling, and Imperfections - First Workshop, DGM4MICCAI 2021, and First Workshop, DALI 2021, Held in Conjunction with MICCAI 2021, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 13003 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 103-111, 1st Workshop on Deep Generative Models for Medical Image Computing and Computer Assisted Intervention, DGM4MICCAI 2021 and 1st Workshop on Data Augmentation, Labelling, and Imperfections, DALI 2021 held in conjunction with 24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021, Virtual, Online, 1/10/21. https://doi.org/10.1007/978-3-030-88210-5_9

Li C, Wei Y, Chen X, Schönlieb CB. BrainNetGAN: Data Augmentation of Brain Connectivity Using Generative Adversarial Network for Dementia Classification. In Engelhardt S, Oksuz I, Zhu D, Yuan Y, Mukhopadhyay A, Heller N, Huang SX, Nguyen H, Sznitman R, Xue Y, editors, Deep Generative Models, and Data Augmentation, Labelling, and Imperfections - First Workshop, DGM4MICCAI 2021, and First Workshop, DALI 2021, Held in Conjunction with MICCAI 2021, Proceedings. Springer Science and Business Media Deutschland GmbH. 2021. p. 103-111. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). Epub 2021 Sept 25. doi: 10.1007/978-3-030-88210-5_9

Li, Chao ; Wei, Yiran ; Chen, Xi et al. / BrainNetGAN : Data Augmentation of Brain Connectivity Using Generative Adversarial Network for Dementia Classification. Deep Generative Models, and Data Augmentation, Labelling, and Imperfections - First Workshop, DGM4MICCAI 2021, and First Workshop, DALI 2021, Held in Conjunction with MICCAI 2021, Proceedings. editor / Sandy Engelhardt ; Ilkay Oksuz ; Dajiang Zhu ; Yixuan Yuan ; Anirban Mukhopadhyay ; Nicholas Heller ; Sharon Xiaolei Huang ; Hien Nguyen ; Raphael Sznitman ; Yuan Xue. Springer Science and Business Media Deutschland GmbH, 2021. pp. 103-111 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{cbe5ec6ef88449fd966c374d12b22671,

title = "BrainNetGAN: Data Augmentation of Brain Connectivity Using Generative Adversarial Network for Dementia Classification",

abstract = "Alzheimer{\textquoteright}s disease (AD) is the most common age-related dementia, which significantly affects an individual{\textquoteright}s daily life and impact socioeconomics. It remains a challenge to identify the individuals at risk of dementia for precise management. Brain MRI offers a non-invasive biomarker to detect brain aging. Previous evidence shows that the structural brain network generated from the diffusion MRI promises to classify dementia accurately based on deep learning models. However, the limited availability of diffusion MRI challenges the model training of deep learning. We propose the BrainNetGAN, a variant of the generative adversarial network, to efficiently augment the structural brain networks for dementia classifying tasks. The BrainNetGAN model is trained to generate fake brain connectivity matrices, which are expected to reflect the latent distribution and topological features of the real brain network data. Numerical results show that the BrainNetGAN outperforms the benchmarking algorithms in augmenting the brain networks for AD classification tasks.",

keywords = "Brain connectivity, Classification, Data augmentation, Generative adversarial network",

author = "Chao Li and Yiran Wei and Xi Chen and Sch{\"o}nlieb, {Carola Bibiane}",

year = "2021",

month = dec,

day = "31",

doi = "10.1007/978-3-030-88210-5_9",

language = "English",

isbn = "9783030882099",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "103--111",

editor = "Sandy Engelhardt and Ilkay Oksuz and Dajiang Zhu and Yixuan Yuan and Anirban Mukhopadhyay and Nicholas Heller and Huang, {Sharon Xiaolei} and Hien Nguyen and Raphael Sznitman and Yuan Xue",

booktitle = "Deep Generative Models, and Data Augmentation, Labelling, and Imperfections - First Workshop, DGM4MICCAI 2021, and First Workshop, DALI 2021, Held in Conjunction with MICCAI 2021, Proceedings",

address = "Germany",

note = "1st Workshop on Deep Generative Models for Medical Image Computing and Computer Assisted Intervention, DGM4MICCAI 2021 and 1st Workshop on Data Augmentation, Labelling, and Imperfections, DALI 2021 held in conjunction with 24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021 ; Conference date: 01-10-2021 Through 01-10-2021",

}

TY - GEN

T1 - BrainNetGAN

T2 - 1st Workshop on Deep Generative Models for Medical Image Computing and Computer Assisted Intervention, DGM4MICCAI 2021 and 1st Workshop on Data Augmentation, Labelling, and Imperfections, DALI 2021 held in conjunction with 24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021

AU - Li, Chao

AU - Wei, Yiran

AU - Chen, Xi

AU - Schönlieb, Carola Bibiane

PY - 2021/12/31

Y1 - 2021/12/31

N2 - Alzheimer’s disease (AD) is the most common age-related dementia, which significantly affects an individual’s daily life and impact socioeconomics. It remains a challenge to identify the individuals at risk of dementia for precise management. Brain MRI offers a non-invasive biomarker to detect brain aging. Previous evidence shows that the structural brain network generated from the diffusion MRI promises to classify dementia accurately based on deep learning models. However, the limited availability of diffusion MRI challenges the model training of deep learning. We propose the BrainNetGAN, a variant of the generative adversarial network, to efficiently augment the structural brain networks for dementia classifying tasks. The BrainNetGAN model is trained to generate fake brain connectivity matrices, which are expected to reflect the latent distribution and topological features of the real brain network data. Numerical results show that the BrainNetGAN outperforms the benchmarking algorithms in augmenting the brain networks for AD classification tasks.

AB - Alzheimer’s disease (AD) is the most common age-related dementia, which significantly affects an individual’s daily life and impact socioeconomics. It remains a challenge to identify the individuals at risk of dementia for precise management. Brain MRI offers a non-invasive biomarker to detect brain aging. Previous evidence shows that the structural brain network generated from the diffusion MRI promises to classify dementia accurately based on deep learning models. However, the limited availability of diffusion MRI challenges the model training of deep learning. We propose the BrainNetGAN, a variant of the generative adversarial network, to efficiently augment the structural brain networks for dementia classifying tasks. The BrainNetGAN model is trained to generate fake brain connectivity matrices, which are expected to reflect the latent distribution and topological features of the real brain network data. Numerical results show that the BrainNetGAN outperforms the benchmarking algorithms in augmenting the brain networks for AD classification tasks.

KW - Brain connectivity

KW - Classification

KW - Data augmentation

KW - Generative adversarial network

UR - http://www.scopus.com/inward/record.url?scp=85116916058&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-88210-5_9

DO - 10.1007/978-3-030-88210-5_9

M3 - Chapter in a published conference proceeding

AN - SCOPUS:85116916058

SN - 9783030882099

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

SP - 103

EP - 111

BT - Deep Generative Models, and Data Augmentation, Labelling, and Imperfections - First Workshop, DGM4MICCAI 2021, and First Workshop, DALI 2021, Held in Conjunction with MICCAI 2021, Proceedings

A2 - Engelhardt, Sandy

A2 - Oksuz, Ilkay

A2 - Zhu, Dajiang

A2 - Yuan, Yixuan

A2 - Mukhopadhyay, Anirban

A2 - Heller, Nicholas

A2 - Huang, Sharon Xiaolei

A2 - Nguyen, Hien

A2 - Sznitman, Raphael

A2 - Xue, Yuan

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 1 October 2021 through 1 October 2021

ER -

BrainNetGAN: Data Augmentation of Brain Connectivity Using Generative Adversarial Network for Dementia Classification

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this