Alzheimer's disease classification using cluster‐based labelling for graph neural network on heterogeneous data

Niamh Mc Combe; Jake Bamrah; Jose Sanchez-Bornot; David Finn; Paula McClean; KongFatt Wong-Lin

doi:10.1049/htl2.12037

Alzheimer's disease classification using cluster‐based labelling for graph neural network on heterogeneous data

Niamh Mc Combe
, Jake Bamrah
, Jose Sanchez-Bornot
, David Finn
, Paula McClean
, KongFatt Wong-Lin

Research output: Contribution to journal › Article › peer-review

9 Citations (Scopus)

122 Downloads (Pure)

Abstract

Biomarkers for Alzheimer's disease (AD) diagnosis do not always correlate reliably with cognitive symptoms, making clinical diagnosis inconsistent. In this study, the performance of a graphical neural network (GNN) classifier based on data‐driven diagnostic classes from unsupervised clustering on heterogeneous data is compared to the performance of a classifier using clinician diagnosis as an outcome. Unsupervised clustering on tau‐positron emission tomography (PET) and cognitive and functional assessment data was performed. Five clusters embedded in a non‐linear uniform manifold approximation and project (UMAP) space were identified. The individual clusters revealed specific feature characteristics with respect to clinical diagnosis of AD, gender, family history, age, and underlying neurological risk factors (NRFs). In particular, one cluster comprised mainly diagnosed AD cases. All cases within this cluster were re‐labelled AD cases. The re‐labelled cases are characterized by high cerebrospinal fluid amyloid beta (CSF Aβ) levels at a younger age, even though Aβ data was not used for clustering. A GNN model was trained using the re‐labelled data with a multiclass area‐under‐the‐curve (AUC) of 95.2%, higher than the AUC of a GNN trained on clinician diagnosis (91.7%; p = 0.02). Overall, our work suggests that more objective cluster‐based diagnostic labels combined with GNN classification may have value in clinical risk stratification and diagnosis of AD.

Original language	English
Pages (from-to)	102-109
Number of pages	8
Journal	Health Technology Letters
Volume	9
Issue number	6
Early online date	20 Oct 2022
DOIs	https://doi.org/10.1049/htl2.12037
Publication status	Published (in print/issue) - 31 Dec 2022

Bibliographical note

Funding Information:
This work was supported by the European Union's INTERREG VA Programme, managed by the Special EU Programmes Body (SEUPB; Centre for Personalised Medicine, IVA 5036). The views and opinions expressed in this paper do not necessarily reflect those of the European Commission or the Special EU Programmes Body (SEUPB). Data collection and sharing for this project was funded by the Alzheimer's Disease Neuroimaging Initiative (ADNI) (National Institutes of Health Grant U01 AG024904) and DOD ADNI (Department of Defense award number W81XWH‐12‐2‐0012). ADNI is funded by the National Institute on Aging, the National Institute of Biomedical Imaging and Bioengineering, and through generous contributions from the following: AbbVie, Alzheimer's Association; Alzheimer's Drug Discovery Foundation; Araclon Biotech; BioClinica, Inc.; Biogen; Bristol‐Myers Squibb Company; CereSpir, Inc.; Cogstate; Eisai Inc.; Elan Pharmaceuticals, Inc.; Eli Lilly and Company; EuroImmun; F. Hoffmann‐La Roche Ltd and its affiliated company Genentech, Inc.; Fujirebio; GE Healthcare; IXICO Ltd.; Janssen Alzheimer Immunotherapy Research & Development, LLC.; Johnson & Johnson Pharmaceutical Research & Development LLC.; Lumosity; Lundbeck; Merck & Co., Inc.; Meso Scale Diagnostics, LLC.; NeuroRx Research; Neurotrack Technologies; Novartis Pharmaceuticals Corporation; Pfizer Inc.; Piramal Imaging; Servier; Takeda Pharmaceutical Company; and Transition Therapeutics. The Canadian Institutes of Health Research is providing funds to support ADNI clinical sites in Canada. Private sector contributions are facilitated by the Foundation for the National Institutes of Health ( www.fnih.org ). The grantee organization is the Northern California Institute for Research and Education, and the study is coordinated by the Alzheimer's Therapeutic Research Institute at the University of Southern California. ADNI data are disseminated by the Laboratory for Neuro Imaging at the University of Southern California.

Publisher Copyright:
© 2022 The Authors. Healthcare Technology Letters published by John Wiley & Sons Ltd on behalf of The Institution of Engineering and Technology.

Funding

Funding Information: This work was supported by the European Union's INTERREG VA Programme, managed by the Special EU Programmes Body (SEUPB; Centre for Personalised Medicine, IVA 5036). The views and opinions expressed in this paper do not necessarily reflect those of the European Commission or the Special EU Programmes Body (SEUPB). Data collection and sharing for this project was funded by the Alzheimer's Disease Neuroimaging Initiative (ADNI) (National Institutes of Health Grant U01 AG024904) and DOD ADNI (Department of Defense award number W81XWH‐12‐2‐0012). ADNI is funded by the National Institute on Aging, the National Institute of Biomedical Imaging and Bioengineering, and through generous contributions from the following: AbbVie, Alzheimer's Association; Alzheimer's Drug Discovery Foundation; Araclon Biotech; BioClinica, Inc.; Biogen; Bristol‐Myers Squibb Company; CereSpir, Inc.; Cogstate; Eisai Inc.; Elan Pharmaceuticals, Inc.; Eli Lilly and Company; EuroImmun; F. Hoffmann‐La Roche Ltd and its affiliated company Genentech, Inc.; Fujirebio; GE Healthcare; IXICO Ltd.; Janssen Alzheimer Immunotherapy Research & Development, LLC.; Johnson & Johnson Pharmaceutical Research & Development LLC.; Lumosity; Lundbeck; Merck & Co., Inc.; Meso Scale Diagnostics, LLC.; NeuroRx Research; Neurotrack Technologies; Novartis Pharmaceuticals Corporation; Pfizer Inc.; Piramal Imaging; Servier; Takeda Pharmaceutical Company; and Transition Therapeutics. The Canadian Institutes of Health Research is providing funds to support ADNI clinical sites in Canada. Private sector contributions are facilitated by the Foundation for the National Institutes of Health ( www.fnih.org ). The grantee organization is the Northern California Institute for Research and Education, and the study is coordinated by the Alzheimer's Therapeutic Research Institute at the University of Southern California. ADNI data are disseminated by the Laboratory for Neuro Imaging at the University of Southern California. Publisher Copyright: © 2022 The Authors. Healthcare Technology Letters published by John Wiley & Sons Ltd on behalf of The Institution of Engineering and Technology.

UN SDGs

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

SDG 3 Good Health and Well-being

Keywords

patient diagnosis
neural nets
learning (artificial intelligence)
feature selection
feature extraction
data visualisation
convolutional neural nets
cognition
brain
biomedical imaging

Access to Document

10.1049/htl2.12037

htl2.12037.pdfFinal published version, 804 KBLicence: CC BY

Cite this

@article{93b18e35817b4cb7ba299654de8feacf,

title = "Alzheimer's disease classification using cluster‐based labelling for graph neural network on heterogeneous data",

abstract = "Biomarkers for Alzheimer's disease (AD) diagnosis do not always correlate reliably with cognitive symptoms, making clinical diagnosis inconsistent. In this study, the performance of a graphical neural network (GNN) classifier based on data‐driven diagnostic classes from unsupervised clustering on heterogeneous data is compared to the performance of a classifier using clinician diagnosis as an outcome. Unsupervised clustering on tau‐positron emission tomography (PET) and cognitive and functional assessment data was performed. Five clusters embedded in a non‐linear uniform manifold approximation and project (UMAP) space were identified. The individual clusters revealed specific feature characteristics with respect to clinical diagnosis of AD, gender, family history, age, and underlying neurological risk factors (NRFs). In particular, one cluster comprised mainly diagnosed AD cases. All cases within this cluster were re‐labelled AD cases. The re‐labelled cases are characterized by high cerebrospinal fluid amyloid beta (CSF Aβ) levels at a younger age, even though Aβ data was not used for clustering. A GNN model was trained using the re‐labelled data with a multiclass area‐under‐the‐curve (AUC) of 95.2\%, higher than the AUC of a GNN trained on clinician diagnosis (91.7\%; p = 0.02). Overall, our work suggests that more objective cluster‐based diagnostic labels combined with GNN classification may have value in clinical risk stratification and diagnosis of AD.",

keywords = "patient diagnosis, neural nets, learning (artificial intelligence), feature selection, feature extraction, data visualisation, convolutional neural nets, cognition, brain, biomedical imaging",

author = "\{Mc Combe\}, Niamh and Jake Bamrah and Jose Sanchez-Bornot and David Finn and Paula McClean and KongFatt Wong-Lin",

note = "Funding Information: This work was supported by the European Union's INTERREG VA Programme, managed by the Special EU Programmes Body (SEUPB; Centre for Personalised Medicine, IVA 5036). The views and opinions expressed in this paper do not necessarily reflect those of the European Commission or the Special EU Programmes Body (SEUPB). Data collection and sharing for this project was funded by the Alzheimer's Disease Neuroimaging Initiative (ADNI) (National Institutes of Health Grant U01 AG024904) and DOD ADNI (Department of Defense award number W81XWH‐12‐2‐0012). ADNI is funded by the National Institute on Aging, the National Institute of Biomedical Imaging and Bioengineering, and through generous contributions from the following: AbbVie, Alzheimer's Association; Alzheimer's Drug Discovery Foundation; Araclon Biotech; BioClinica, Inc.; Biogen; Bristol‐Myers Squibb Company; CereSpir, Inc.; Cogstate; Eisai Inc.; Elan Pharmaceuticals, Inc.; Eli Lilly and Company; EuroImmun; F. Hoffmann‐La Roche Ltd and its affiliated company Genentech, Inc.; Fujirebio; GE Healthcare; IXICO Ltd.; Janssen Alzheimer Immunotherapy Research \& Development, LLC.; Johnson \& Johnson Pharmaceutical Research \& Development LLC.; Lumosity; Lundbeck; Merck \& Co., Inc.; Meso Scale Diagnostics, LLC.; NeuroRx Research; Neurotrack Technologies; Novartis Pharmaceuticals Corporation; Pfizer Inc.; Piramal Imaging; Servier; Takeda Pharmaceutical Company; and Transition Therapeutics. The Canadian Institutes of Health Research is providing funds to support ADNI clinical sites in Canada. Private sector contributions are facilitated by the Foundation for the National Institutes of Health ( www.fnih.org ). The grantee organization is the Northern California Institute for Research and Education, and the study is coordinated by the Alzheimer's Therapeutic Research Institute at the University of Southern California. ADNI data are disseminated by the Laboratory for Neuro Imaging at the University of Southern California. Publisher Copyright: {\textcopyright} 2022 The Authors. Healthcare Technology Letters published by John Wiley \& Sons Ltd on behalf of The Institution of Engineering and Technology.",

year = "2022",

month = dec,

day = "31",

doi = "10.1049/htl2.12037",

language = "English",

volume = "9",

pages = "102--109",

journal = "Health Technology Letters",

issn = "2053-3713",

publisher = "John Wiley \& Sons Inc.",

number = "6",

}

TY - JOUR

T1 - Alzheimer's disease classification using cluster‐based labelling for graph neural network on heterogeneous data

AU - Mc Combe, Niamh

AU - Bamrah, Jake

AU - Sanchez-Bornot, Jose

AU - Finn, David

AU - McClean, Paula

AU - Wong-Lin, KongFatt

N1 - Funding Information: This work was supported by the European Union's INTERREG VA Programme, managed by the Special EU Programmes Body (SEUPB; Centre for Personalised Medicine, IVA 5036). The views and opinions expressed in this paper do not necessarily reflect those of the European Commission or the Special EU Programmes Body (SEUPB). Data collection and sharing for this project was funded by the Alzheimer's Disease Neuroimaging Initiative (ADNI) (National Institutes of Health Grant U01 AG024904) and DOD ADNI (Department of Defense award number W81XWH‐12‐2‐0012). ADNI is funded by the National Institute on Aging, the National Institute of Biomedical Imaging and Bioengineering, and through generous contributions from the following: AbbVie, Alzheimer's Association; Alzheimer's Drug Discovery Foundation; Araclon Biotech; BioClinica, Inc.; Biogen; Bristol‐Myers Squibb Company; CereSpir, Inc.; Cogstate; Eisai Inc.; Elan Pharmaceuticals, Inc.; Eli Lilly and Company; EuroImmun; F. Hoffmann‐La Roche Ltd and its affiliated company Genentech, Inc.; Fujirebio; GE Healthcare; IXICO Ltd.; Janssen Alzheimer Immunotherapy Research & Development, LLC.; Johnson & Johnson Pharmaceutical Research & Development LLC.; Lumosity; Lundbeck; Merck & Co., Inc.; Meso Scale Diagnostics, LLC.; NeuroRx Research; Neurotrack Technologies; Novartis Pharmaceuticals Corporation; Pfizer Inc.; Piramal Imaging; Servier; Takeda Pharmaceutical Company; and Transition Therapeutics. The Canadian Institutes of Health Research is providing funds to support ADNI clinical sites in Canada. Private sector contributions are facilitated by the Foundation for the National Institutes of Health ( www.fnih.org ). The grantee organization is the Northern California Institute for Research and Education, and the study is coordinated by the Alzheimer's Therapeutic Research Institute at the University of Southern California. ADNI data are disseminated by the Laboratory for Neuro Imaging at the University of Southern California. Publisher Copyright: © 2022 The Authors. Healthcare Technology Letters published by John Wiley & Sons Ltd on behalf of The Institution of Engineering and Technology.

PY - 2022/12/31

Y1 - 2022/12/31

N2 - Biomarkers for Alzheimer's disease (AD) diagnosis do not always correlate reliably with cognitive symptoms, making clinical diagnosis inconsistent. In this study, the performance of a graphical neural network (GNN) classifier based on data‐driven diagnostic classes from unsupervised clustering on heterogeneous data is compared to the performance of a classifier using clinician diagnosis as an outcome. Unsupervised clustering on tau‐positron emission tomography (PET) and cognitive and functional assessment data was performed. Five clusters embedded in a non‐linear uniform manifold approximation and project (UMAP) space were identified. The individual clusters revealed specific feature characteristics with respect to clinical diagnosis of AD, gender, family history, age, and underlying neurological risk factors (NRFs). In particular, one cluster comprised mainly diagnosed AD cases. All cases within this cluster were re‐labelled AD cases. The re‐labelled cases are characterized by high cerebrospinal fluid amyloid beta (CSF Aβ) levels at a younger age, even though Aβ data was not used for clustering. A GNN model was trained using the re‐labelled data with a multiclass area‐under‐the‐curve (AUC) of 95.2%, higher than the AUC of a GNN trained on clinician diagnosis (91.7%; p = 0.02). Overall, our work suggests that more objective cluster‐based diagnostic labels combined with GNN classification may have value in clinical risk stratification and diagnosis of AD.

AB - Biomarkers for Alzheimer's disease (AD) diagnosis do not always correlate reliably with cognitive symptoms, making clinical diagnosis inconsistent. In this study, the performance of a graphical neural network (GNN) classifier based on data‐driven diagnostic classes from unsupervised clustering on heterogeneous data is compared to the performance of a classifier using clinician diagnosis as an outcome. Unsupervised clustering on tau‐positron emission tomography (PET) and cognitive and functional assessment data was performed. Five clusters embedded in a non‐linear uniform manifold approximation and project (UMAP) space were identified. The individual clusters revealed specific feature characteristics with respect to clinical diagnosis of AD, gender, family history, age, and underlying neurological risk factors (NRFs). In particular, one cluster comprised mainly diagnosed AD cases. All cases within this cluster were re‐labelled AD cases. The re‐labelled cases are characterized by high cerebrospinal fluid amyloid beta (CSF Aβ) levels at a younger age, even though Aβ data was not used for clustering. A GNN model was trained using the re‐labelled data with a multiclass area‐under‐the‐curve (AUC) of 95.2%, higher than the AUC of a GNN trained on clinician diagnosis (91.7%; p = 0.02). Overall, our work suggests that more objective cluster‐based diagnostic labels combined with GNN classification may have value in clinical risk stratification and diagnosis of AD.

KW - patient diagnosis

KW - neural nets

KW - learning (artificial intelligence)

KW - feature selection

KW - feature extraction

KW - data visualisation

KW - convolutional neural nets

KW - cognition

KW - brain

KW - biomedical imaging

UR - https://www.scopus.com/pages/publications/85140249529

U2 - 10.1049/htl2.12037

DO - 10.1049/htl2.12037

M3 - Article

C2 - 36514476

SN - 2053-3713

VL - 9

SP - 102

EP - 109

JO - Health Technology Letters

JF - Health Technology Letters

IS - 6

ER -

Alzheimer's disease classification using cluster‐based labelling for graph neural network on heterogeneous data

Abstract

Bibliographical note

Funding

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this