EEG-GMACN: Interpretable EEG Graph Mutual Attention Convolutional Network

Haili Ye; Stephan Goerttler; Fei He

doi:10.1109/EMBC53108.2024.10782694

EEG-GMACN: Interpretable EEG Graph Mutual Attention Convolutional Network

Haili Ye, Stephan Goerttler, Fei He

Research Centre for Computational Science and Mathematical Modelling

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding › peer-review

1 Downloads (Pure)

Abstract

Electroencephalogram (EEG) is a valuable technique to record brain electrical activity through electrodes placed on the scalp. Analyzing EEG signals contributes to the understanding of neurological conditions and developing brain-computer interface. Graph Signal Processing (GSP) has emerged as a promising method for EEG spatial-temporal analysis, by further considering the topological relationships between electrodes. However, existing GSP studies lack interpretability of electrode importance and the credibility of prediction confidence. This work proposes an EEG Graph Mutual Attention Convolutional Network (EEG-GMACN), by introducing an ’Inverse Graph Weight Module’ to output interpretable electrode graph weights, enhancing the clinical credibility and interpretability of EEG classification results. Additionally, we incorporate a mutual attention mechanism module into the model to improve its capability to distinguish critical electrodes and introduce credibility calibration to assess the uncertainty of prediction results. This study enhances the transparency and effectiveness of EEG analysis, paving the way for its widespread use in clinical and neuroscience research.

Original language	English
Title of host publication	2024 46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)
Publisher	IEEE
Pages	1-4
Number of pages	4
ISBN (Electronic)	9798350371499
ISBN (Print)	9798350371505
DOIs	https://doi.org/10.1109/EMBC53108.2024.10782694
Publication status	E-pub ahead of print - 17 Dec 2024
Event	46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society - Orlando, FL, USA, Orlando, United States Duration: 15 Jul 2024 → 19 Jul 2024 https://embc.embs.org/2024/

Publication series

Name	Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS
ISSN (Print)	1557-170X

Conference

Conference	46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society
Abbreviated title	IEEE EMBS
Country/Territory	United States
City	Orlando
Period	15/07/24 → 19/07/24
Internet address	https://embc.embs.org/2024/

Bibliographical note

This document is the author’s post-print version, incorporating any revisions agreed during the peer-review process. Some differences between the published version and this version may remain and you are advised to consult the published version if you wish to cite from it.

Funding

This work is in part supported by the Engineering and Physical Sciences Research Council (EPSRC) under Grant EP/X020193/1.

Funders	Funder number
Engineering and Physical Sciences Research Council	EP/X020193/1
Engineering and Physical Sciences Research Council

Keywords

Electrodes
Neuroscience
Uncertainty
Convolution
Scalp
Predictive models
Brain modeling
Electroencephalography
Convolutional neural networks
Engineering in medicine and biology

Access to Document

10.1109/EMBC53108.2024.10782694

He2024VORAccepted author manuscript, 825 KB

Cite this

Ye, H., Goerttler, S., & He, F. (2024). EEG-GMACN: Interpretable EEG Graph Mutual Attention Convolutional Network. In 2024 46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) (pp. 1-4). Article 10782694 (Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS). IEEE. Advance online publication. https://doi.org/10.1109/EMBC53108.2024.10782694

EEG-GMACN: Interpretable EEG Graph Mutual Attention Convolutional Network. / Ye, Haili ; Goerttler, Stephan ; He, Fei.
2024 46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). IEEE, 2024. p. 1-4 10782694 (Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS).

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding › peer-review

Ye, H , Goerttler, S & He, F 2024, EEG-GMACN: Interpretable EEG Graph Mutual Attention Convolutional Network. in 2024 46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)., 10782694, Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, IEEE, pp. 1-4, 46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Orlando, Florida, United States, 15/07/24. https://doi.org/10.1109/EMBC53108.2024.10782694

Ye H , Goerttler S , He F. EEG-GMACN: Interpretable EEG Graph Mutual Attention Convolutional Network. In 2024 46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). IEEE. 2024. p. 1-4. 10782694. (Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS). Epub 2024 Dec 17. doi: 10.1109/EMBC53108.2024.10782694

@inproceedings{980f4052f1a349bcb8bfea340382a972,

title = "EEG-GMACN: Interpretable EEG Graph Mutual Attention Convolutional Network",

abstract = "Electroencephalogram (EEG) is a valuable technique to record brain electrical activity through electrodes placed on the scalp. Analyzing EEG signals contributes to the understanding of neurological conditions and developing brain-computer interface. Graph Signal Processing (GSP) has emerged as a promising method for EEG spatial-temporal analysis, by further considering the topological relationships between electrodes. However, existing GSP studies lack interpretability of electrode importance and the credibility of prediction confidence. This work proposes an EEG Graph Mutual Attention Convolutional Network (EEG-GMACN), by introducing an {\textquoteright}Inverse Graph Weight Module{\textquoteright} to output interpretable electrode graph weights, enhancing the clinical credibility and interpretability of EEG classification results. Additionally, we incorporate a mutual attention mechanism module into the model to improve its capability to distinguish critical electrodes and introduce credibility calibration to assess the uncertainty of prediction results. This study enhances the transparency and effectiveness of EEG analysis, paving the way for its widespread use in clinical and neuroscience research.",

keywords = "Electrodes, Neuroscience, Uncertainty, Convolution, Scalp, Predictive models, Brain modeling, Electroencephalography, Convolutional neural networks, Engineering in medicine and biology",

author = "Haili Ye and Stephan Goerttler and Fei He",

note = "This document is the author{\textquoteright}s post-print version, incorporating any revisions agreed during the peer-review process. Some differences between the published version and this version may remain and you are advised to consult the published version if you wish to cite from it.; 46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, IEEE EMBS ; Conference date: 15-07-2024 Through 19-07-2024",

year = "2024",

month = dec,

day = "17",

doi = "10.1109/EMBC53108.2024.10782694",

language = "English",

isbn = "9798350371505",

series = "Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS",

publisher = "IEEE",

pages = "1--4",

booktitle = "2024 46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)",

address = "United States",

url = "https://embc.embs.org/2024/",

}

TY - GEN

T1 - EEG-GMACN

T2 - 46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society

AU - Ye, Haili

AU - Goerttler, Stephan

AU - He, Fei

N1 - This document is the author’s post-print version, incorporating any revisions agreed during the peer-review process. Some differences between the published version and this version may remain and you are advised to consult the published version if you wish to cite from it.

PY - 2024/12/17

Y1 - 2024/12/17

N2 - Electroencephalogram (EEG) is a valuable technique to record brain electrical activity through electrodes placed on the scalp. Analyzing EEG signals contributes to the understanding of neurological conditions and developing brain-computer interface. Graph Signal Processing (GSP) has emerged as a promising method for EEG spatial-temporal analysis, by further considering the topological relationships between electrodes. However, existing GSP studies lack interpretability of electrode importance and the credibility of prediction confidence. This work proposes an EEG Graph Mutual Attention Convolutional Network (EEG-GMACN), by introducing an ’Inverse Graph Weight Module’ to output interpretable electrode graph weights, enhancing the clinical credibility and interpretability of EEG classification results. Additionally, we incorporate a mutual attention mechanism module into the model to improve its capability to distinguish critical electrodes and introduce credibility calibration to assess the uncertainty of prediction results. This study enhances the transparency and effectiveness of EEG analysis, paving the way for its widespread use in clinical and neuroscience research.

AB - Electroencephalogram (EEG) is a valuable technique to record brain electrical activity through electrodes placed on the scalp. Analyzing EEG signals contributes to the understanding of neurological conditions and developing brain-computer interface. Graph Signal Processing (GSP) has emerged as a promising method for EEG spatial-temporal analysis, by further considering the topological relationships between electrodes. However, existing GSP studies lack interpretability of electrode importance and the credibility of prediction confidence. This work proposes an EEG Graph Mutual Attention Convolutional Network (EEG-GMACN), by introducing an ’Inverse Graph Weight Module’ to output interpretable electrode graph weights, enhancing the clinical credibility and interpretability of EEG classification results. Additionally, we incorporate a mutual attention mechanism module into the model to improve its capability to distinguish critical electrodes and introduce credibility calibration to assess the uncertainty of prediction results. This study enhances the transparency and effectiveness of EEG analysis, paving the way for its widespread use in clinical and neuroscience research.

KW - Electrodes

KW - Neuroscience

KW - Uncertainty

KW - Convolution

KW - Scalp

KW - Predictive models

KW - Brain modeling

KW - Electroencephalography

KW - Convolutional neural networks

KW - Engineering in medicine and biology

U2 - 10.1109/EMBC53108.2024.10782694

DO - 10.1109/EMBC53108.2024.10782694

M3 - Conference proceeding

SN - 9798350371505

T3 - Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS

SP - 1

EP - 4

BT - 2024 46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)

PB - IEEE

Y2 - 15 July 2024 through 19 July 2024

ER -

EEG-GMACN: Interpretable EEG Graph Mutual Attention Convolutional Network

Abstract

Publication series

Conference

Bibliographical note

Funding

Keywords

Access to Document

Fingerprint

Cite this