Graph-BrainConvNet: A One-class GCN-based approach for MCI detection from source-level MEG

Kaniska Samanta; Jose Sanchez Bornot; Paula McClean; Daman Kaur; Girijesh Prasad; Saugat Bhattacharyya; KongFatt Wong-Lin

doi:10.1109/EMBC58623.2025.11254716

Graph-BrainConvNet: A One-class GCN-based approach for MCI detection from source-level MEG

Research output: Contribution to conference › Paper › peer-review

Abstract

—Mild cognitive impairment (MCI) is a neurocognitive disorder that increases the risk of dementia in older age, making early detection crucial for subsequent risk reduction strategies. Existing studies with advanced neuroimaging techniques mostly focused on finding MCI-related neuromarkers either during resting state (without evaluating their cognitive ability) or during complex cognitive tasks, which can be challenging for non-MCI geriatrics as well. This paper introduces a novel feasibility experiment to identify neural correlates related to attention from MCI patients’ while watching a silent movie with random auditory distractions. A novel one-class graph convolutional neural network (GCN)-based model is proposed to detect MCI neuromarkers using source-level functional brain connectivity and spectral-temporal features, achieving a classif ication accuracy of 92.63±0.96% between control participants (not diagnosed with MCI) and MCI for the gamma rhythm. MCI patients showed higher responses to auditory distractors than control participants (CPs), indicating cognitive decline. Our experiment reveals potential auditory gating deficits in MCI, which the proposed GCN model can capture for early diagnosis.

Original language	English
Pages	1-7
Number of pages	7
DOIs	https://doi.org/10.1109/EMBC58623.2025.11254716
Publication status	Accepted/In press - 8 Apr 2025
Event	47th Annual International Conference of the IEEE Engineering in Medicine and Biology Society - Copenhagen, Denmark Duration: 14 Jul 2025 → … https://embc.embs.org/2025/

Conference

Conference	47th Annual International Conference of the IEEE Engineering in Medicine and Biology Society
Abbreviated title	EMBC 2025
Country/Territory	Denmark
City	Copenhagen
Period	14/07/25 → …
Internet address	https://embc.embs.org/2025/

Funding

K.S. was funded by Ulster University’s Vice Chancellor Research Scholarship. The acquisition and collation of the experimental data was supported by the European Union’s INTERREG VA Programme, managed by the Special EU Programmes (SEUPB; Centre for Personalised Medicine, IVA 5036) (J.M.S.-B., D.K., G.P., P.L.M., K.W.-L.), with additional support by the Northern Ireland Functional Brain Mapping Project Facility (1303/101154803) and funded by Invest Northern Ireland and the University of Ulster (G.P., K.W.-L.). The views and opinions expressed in this article do not necessarily reflect those of the European Commission or the Special EU Programmes Body (SEUPB). We are grateful for access to the Tier 2 High Performance Computing resources provided by the Northern Ireland High Performance Computing (NI-HPC) facility funded by the UK Engineering and Physical Sciences Research Council (EPSRC), Grant No. EP/T022175/1.

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being
SDG 4 Quality Education
SDG 10 Reduced Inequalities

Keywords

Mild Cognitive Impairment
Graph Convolutional Network
magnetoencephalography
source localization
linearly constrained minimumvariance (LCMV) beamforming
functional brain connectivity
imaginary part of coherency
Neural Networks, Computer
Humans
Cognitive Dysfunction/diagnosis
Male
Brain/physiopathology
Magnetoencephalography/methods
Female
Aged

Access to Document

10.1109/EMBC58623.2025.11254716

Cite this

Samanta, K., Sanchez Bornot, J., McClean, P., Kaur, D., Prasad, G., Bhattacharyya, S., & Wong-Lin, K. (Accepted/In press). Graph-BrainConvNet: A One-class GCN-based approach for MCI detection from source-level MEG. 1-7. Paper presented at 47th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Copenhagen, Denmark. https://doi.org/10.1109/EMBC58623.2025.11254716

@conference{7569138df4dd4281b6c2279d1455d448,

title = "Graph-BrainConvNet: A One-class GCN-based approach for MCI detection from source-level MEG",

abstract = "—Mild cognitive impairment (MCI) is a neurocognitive disorder that increases the risk of dementia in older age, making early detection crucial for subsequent risk reduction strategies. Existing studies with advanced neuroimaging techniques mostly focused on finding MCI-related neuromarkers either during resting state (without evaluating their cognitive ability) or during complex cognitive tasks, which can be challenging for non-MCI geriatrics as well. This paper introduces a novel feasibility experiment to identify neural correlates related to attention from MCI patients{\textquoteright} while watching a silent movie with random auditory distractions. A novel one-class graph convolutional neural network (GCN)-based model is proposed to detect MCI neuromarkers using source-level functional brain connectivity and spectral-temporal features, achieving a classif ication accuracy of 92.63±0.96\% between control participants (not diagnosed with MCI) and MCI for the gamma rhythm. MCI patients showed higher responses to auditory distractors than control participants (CPs), indicating cognitive decline. Our experiment reveals potential auditory gating deficits in MCI, which the proposed GCN model can capture for early diagnosis.",

keywords = "Mild Cognitive Impairment, Graph Convolutional Network, magnetoencephalography, source localization, linearly constrained minimumvariance (LCMV) beamforming, functional brain connectivity, imaginary part of coherency, Neural Networks, Computer, Humans, Cognitive Dysfunction/diagnosis, Male, Brain/physiopathology, Magnetoencephalography/methods, Female, Aged",

author = "Kaniska Samanta and \{Sanchez Bornot\}, Jose and Paula McClean and Daman Kaur and Girijesh Prasad and Saugat Bhattacharyya and KongFatt Wong-Lin",

year = "2025",

month = apr,

day = "8",

doi = "10.1109/EMBC58623.2025.11254716",

language = "English",

pages = "1--7",

note = "47th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2025 ; Conference date: 14-07-2025",

url = "https://embc.embs.org/2025/",

}

Samanta, K , Sanchez Bornot, J , McClean, P, Kaur, D, Prasad, G , Bhattacharyya, S & Wong-Lin, K 2025, 'Graph-BrainConvNet: A One-class GCN-based approach for MCI detection from source-level MEG', Paper presented at 47th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Copenhagen, Denmark, 14/07/25 pp. 1-7. https://doi.org/10.1109/EMBC58623.2025.11254716

TY - CONF

T1 - Graph-BrainConvNet: A One-class GCN-based approach for MCI detection from source-level MEG

AU - Samanta, Kaniska

AU - Sanchez Bornot, Jose

AU - McClean, Paula

AU - Kaur, Daman

AU - Prasad, Girijesh

AU - Bhattacharyya, Saugat

AU - Wong-Lin, KongFatt

PY - 2025/4/8

Y1 - 2025/4/8

N2 - —Mild cognitive impairment (MCI) is a neurocognitive disorder that increases the risk of dementia in older age, making early detection crucial for subsequent risk reduction strategies. Existing studies with advanced neuroimaging techniques mostly focused on finding MCI-related neuromarkers either during resting state (without evaluating their cognitive ability) or during complex cognitive tasks, which can be challenging for non-MCI geriatrics as well. This paper introduces a novel feasibility experiment to identify neural correlates related to attention from MCI patients’ while watching a silent movie with random auditory distractions. A novel one-class graph convolutional neural network (GCN)-based model is proposed to detect MCI neuromarkers using source-level functional brain connectivity and spectral-temporal features, achieving a classif ication accuracy of 92.63±0.96% between control participants (not diagnosed with MCI) and MCI for the gamma rhythm. MCI patients showed higher responses to auditory distractors than control participants (CPs), indicating cognitive decline. Our experiment reveals potential auditory gating deficits in MCI, which the proposed GCN model can capture for early diagnosis.

AB - —Mild cognitive impairment (MCI) is a neurocognitive disorder that increases the risk of dementia in older age, making early detection crucial for subsequent risk reduction strategies. Existing studies with advanced neuroimaging techniques mostly focused on finding MCI-related neuromarkers either during resting state (without evaluating their cognitive ability) or during complex cognitive tasks, which can be challenging for non-MCI geriatrics as well. This paper introduces a novel feasibility experiment to identify neural correlates related to attention from MCI patients’ while watching a silent movie with random auditory distractions. A novel one-class graph convolutional neural network (GCN)-based model is proposed to detect MCI neuromarkers using source-level functional brain connectivity and spectral-temporal features, achieving a classif ication accuracy of 92.63±0.96% between control participants (not diagnosed with MCI) and MCI for the gamma rhythm. MCI patients showed higher responses to auditory distractors than control participants (CPs), indicating cognitive decline. Our experiment reveals potential auditory gating deficits in MCI, which the proposed GCN model can capture for early diagnosis.

KW - Mild Cognitive Impairment

KW - Graph Convolutional Network

KW - magnetoencephalography

KW - source localization

KW - linearly constrained minimumvariance (LCMV) beamforming

KW - functional brain connectivity

KW - imaginary part of coherency

KW - Neural Networks, Computer

KW - Humans

KW - Cognitive Dysfunction/diagnosis

KW - Male

KW - Brain/physiopathology

KW - Magnetoencephalography/methods

KW - Female

KW - Aged

UR - https://www.scopus.com/pages/publications/105023727409

U2 - 10.1109/EMBC58623.2025.11254716

DO - 10.1109/EMBC58623.2025.11254716

M3 - Paper

C2 - 41337313

SP - 1

EP - 7

T2 - 47th Annual International Conference of the IEEE Engineering in Medicine and Biology Society

Y2 - 14 July 2025

ER -

Graph-BrainConvNet: A One-class GCN-based approach for MCI detection from source-level MEG

Abstract

Conference

Funding

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Integrative biology and stratification of dementia

Cite this

Graph-BrainConvNet: A One-class GCN-based approach for MCI detection from source-level MEG

Abstract

Conference

Funding

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Student theses

Integrative biology and stratification of dementia

Cite this