A weighted patient network-based framework for predicting chronic diseases using graph neural networks

Abstract Chronic disease prediction is a critical task in healthcare. Existing studies fulfil this requirement by employing machine learning techniques based on patient features, but they suffer from high dimensional data problems and a high level of bias. We propose a framework for predicting chron...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Haohui Lu, Shahadat Uddin
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/c10890ff33224ca0be3b1354b803e761
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:c10890ff33224ca0be3b1354b803e761
record_format	dspace
spelling	oai:doaj.org-article:c10890ff33224ca0be3b1354b803e7612021-11-21T12:19:54ZA weighted patient network-based framework for predicting chronic diseases using graph neural networks10.1038/s41598-021-01964-22045-2322https://doaj.org/article/c10890ff33224ca0be3b1354b803e7612021-11-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-01964-2https://doaj.org/toc/2045-2322Abstract Chronic disease prediction is a critical task in healthcare. Existing studies fulfil this requirement by employing machine learning techniques based on patient features, but they suffer from high dimensional data problems and a high level of bias. We propose a framework for predicting chronic disease based on Graph Neural Networks (GNNs) to address these issues. We begin by projecting a patient-disease bipartite graph to create a weighted patient network (WPN) that extracts the latent relationship among patients. We then use GNN-based techniques to build prediction models. These models use features extracted from WPN to create robust patient representations for chronic disease prediction. We compare the output of GNN-based models to machine learning methods by using cardiovascular disease and chronic pulmonary disease. The results show that our framework enhances the accuracy of chronic disease prediction. The model with attention mechanisms achieves an accuracy of 93.49% for cardiovascular disease prediction and 89.15% for chronic pulmonary disease prediction. Furthermore, the visualisation of the last hidden layers of GNN-based models shows the pattern for the two cohorts, demonstrating the discriminative strength of the framework. The proposed framework can help stakeholders improve health management systems for patients at risk of developing chronic diseases and conditions.Haohui LuShahadat UddinNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-12 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Haohui Lu Shahadat Uddin A weighted patient network-based framework for predicting chronic diseases using graph neural networks
description	Abstract Chronic disease prediction is a critical task in healthcare. Existing studies fulfil this requirement by employing machine learning techniques based on patient features, but they suffer from high dimensional data problems and a high level of bias. We propose a framework for predicting chronic disease based on Graph Neural Networks (GNNs) to address these issues. We begin by projecting a patient-disease bipartite graph to create a weighted patient network (WPN) that extracts the latent relationship among patients. We then use GNN-based techniques to build prediction models. These models use features extracted from WPN to create robust patient representations for chronic disease prediction. We compare the output of GNN-based models to machine learning methods by using cardiovascular disease and chronic pulmonary disease. The results show that our framework enhances the accuracy of chronic disease prediction. The model with attention mechanisms achieves an accuracy of 93.49% for cardiovascular disease prediction and 89.15% for chronic pulmonary disease prediction. Furthermore, the visualisation of the last hidden layers of GNN-based models shows the pattern for the two cohorts, demonstrating the discriminative strength of the framework. The proposed framework can help stakeholders improve health management systems for patients at risk of developing chronic diseases and conditions.
format	article
author	Haohui Lu Shahadat Uddin
author_facet	Haohui Lu Shahadat Uddin
author_sort	Haohui Lu
title	A weighted patient network-based framework for predicting chronic diseases using graph neural networks
title_short	A weighted patient network-based framework for predicting chronic diseases using graph neural networks
title_full	A weighted patient network-based framework for predicting chronic diseases using graph neural networks
title_fullStr	A weighted patient network-based framework for predicting chronic diseases using graph neural networks
title_full_unstemmed	A weighted patient network-based framework for predicting chronic diseases using graph neural networks
title_sort	weighted patient network-based framework for predicting chronic diseases using graph neural networks
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/c10890ff33224ca0be3b1354b803e761
work_keys_str_mv	AT haohuilu aweightedpatientnetworkbasedframeworkforpredictingchronicdiseasesusinggraphneuralnetworks AT shahadatuddin aweightedpatientnetworkbasedframeworkforpredictingchronicdiseasesusinggraphneuralnetworks AT haohuilu weightedpatientnetworkbasedframeworkforpredictingchronicdiseasesusinggraphneuralnetworks AT shahadatuddin weightedpatientnetworkbasedframeworkforpredictingchronicdiseasesusinggraphneuralnetworks
_version_	1718419091681705984

A weighted patient network-based framework for predicting chronic diseases using graph neural networks

Ejemplares similares