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 cl...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Wiley
2022-12-01
|
Series: | Healthcare Technology Letters |
Online Access: | https://doi.org/10.1049/htl2.12037 |
_version_ | 1811178347672109056 |
---|---|
author | Niamh McCombe Jake Bamrah Jose M. Sanchez‐Bornot David P. Finn Paula L. McClean KongFatt Wong‐Lin Alzheimer's Disease Neuroimaging Initiative (ADNI) |
author_facet | Niamh McCombe Jake Bamrah Jose M. Sanchez‐Bornot David P. Finn Paula L. McClean KongFatt Wong‐Lin Alzheimer's Disease Neuroimaging Initiative (ADNI) |
author_sort | Niamh McCombe |
collection | DOAJ |
description | 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. |
first_indexed | 2024-04-11T06:17:53Z |
format | Article |
id | doaj.art-d2adecc6ed4a41499ab50a32cb7c79ca |
institution | Directory Open Access Journal |
issn | 2053-3713 |
language | English |
last_indexed | 2024-04-11T06:17:53Z |
publishDate | 2022-12-01 |
publisher | Wiley |
record_format | Article |
series | Healthcare Technology Letters |
spelling | doaj.art-d2adecc6ed4a41499ab50a32cb7c79ca2022-12-22T04:41:01ZengWileyHealthcare Technology Letters2053-37132022-12-019610210910.1049/htl2.12037Alzheimer's disease classification using cluster‐based labelling for graph neural network on heterogeneous dataNiamh McCombe0Jake Bamrah1Jose M. Sanchez‐Bornot2David P. Finn3Paula L. McClean4KongFatt Wong‐Lin5Alzheimer's Disease Neuroimaging Initiative (ADNI)Intelligent Systems Research Centre School of Computing Engineering and Intelligent Systems Ulster University Derry∼Londonderry Northern Ireland UKIntelligent Systems Research Centre School of Computing Engineering and Intelligent Systems Ulster University Derry∼Londonderry Northern Ireland UKIntelligent Systems Research Centre School of Computing Engineering and Intelligent Systems Ulster University Derry∼Londonderry Northern Ireland UKPharmacology and Therapeutics, Galway Neuroscience Centre, Centre for Pain Research, and School of Medicine National University of Ireland Galway Galway IrelandNorthern Ireland Centre for Stratified Medicine, Biomedical Sciences Research Institute, Clinical Translational Research and Innovation Centre (C‐TRIC) Ulster University Derry∼Londonderry Northern Ireland UKIntelligent Systems Research Centre School of Computing Engineering and Intelligent Systems Ulster University Derry∼Londonderry Northern Ireland UKAbstract 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.https://doi.org/10.1049/htl2.12037 |
spellingShingle | Niamh McCombe Jake Bamrah Jose M. Sanchez‐Bornot David P. Finn Paula L. McClean KongFatt Wong‐Lin Alzheimer's Disease Neuroimaging Initiative (ADNI) Alzheimer's disease classification using cluster‐based labelling for graph neural network on heterogeneous data Healthcare Technology Letters |
title | Alzheimer's disease classification using cluster‐based labelling for graph neural network on heterogeneous data |
title_full | Alzheimer's disease classification using cluster‐based labelling for graph neural network on heterogeneous data |
title_fullStr | Alzheimer's disease classification using cluster‐based labelling for graph neural network on heterogeneous data |
title_full_unstemmed | Alzheimer's disease classification using cluster‐based labelling for graph neural network on heterogeneous data |
title_short | Alzheimer's disease classification using cluster‐based labelling for graph neural network on heterogeneous data |
title_sort | alzheimer s disease classification using cluster based labelling for graph neural network on heterogeneous data |
url | https://doi.org/10.1049/htl2.12037 |
work_keys_str_mv | AT niamhmccombe alzheimersdiseaseclassificationusingclusterbasedlabellingforgraphneuralnetworkonheterogeneousdata AT jakebamrah alzheimersdiseaseclassificationusingclusterbasedlabellingforgraphneuralnetworkonheterogeneousdata AT josemsanchezbornot alzheimersdiseaseclassificationusingclusterbasedlabellingforgraphneuralnetworkonheterogeneousdata AT davidpfinn alzheimersdiseaseclassificationusingclusterbasedlabellingforgraphneuralnetworkonheterogeneousdata AT paulalmcclean alzheimersdiseaseclassificationusingclusterbasedlabellingforgraphneuralnetworkonheterogeneousdata AT kongfattwonglin alzheimersdiseaseclassificationusingclusterbasedlabellingforgraphneuralnetworkonheterogeneousdata AT alzheimersdiseaseneuroimaginginitiativeadni alzheimersdiseaseclassificationusingclusterbasedlabellingforgraphneuralnetworkonheterogeneousdata |