Comparison of machine learning methods for estimating case fatality ratios: An Ebola outbreak simulation study.

<h4>Background</h4>Machine learning (ML) algorithms are now increasingly used in infectious disease epidemiology. Epidemiologists should understand how ML algorithms behave within the context of outbreak data where missingness of data is almost ubiquitous.<h4>Methods</h4>Usin...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Alpha Forna, Ilaria Dorigatti, Pierre Nouvellet, Christl A Donnelly
Formato:	article
Lenguaje:	EN
Publicado:	Public Library of Science (PLoS) 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/f76e19b1b9e84c1795c43352c11e6790
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:f76e19b1b9e84c1795c43352c11e6790
record_format	dspace
spelling	oai:doaj.org-article:f76e19b1b9e84c1795c43352c11e67902021-12-02T20:14:37ZComparison of machine learning methods for estimating case fatality ratios: An Ebola outbreak simulation study.1932-620310.1371/journal.pone.0257005https://doaj.org/article/f76e19b1b9e84c1795c43352c11e67902021-01-01T00:00:00Zhttps://doi.org/10.1371/journal.pone.0257005https://doaj.org/toc/1932-6203<h4>Background</h4>Machine learning (ML) algorithms are now increasingly used in infectious disease epidemiology. Epidemiologists should understand how ML algorithms behave within the context of outbreak data where missingness of data is almost ubiquitous.<h4>Methods</h4>Using simulated data, we use a ML algorithmic framework to evaluate data imputation performance and the resulting case fatality ratio (CFR) estimates, focusing on the scale and type of data missingness (i.e., missing completely at random-MCAR, missing at random-MAR, or missing not at random-MNAR).<h4>Results</h4>Across ML methods, dataset sizes and proportions of training data used, the area under the receiver operating characteristic curve decreased by 7% (median, range: 1%-16%) when missingness was increased from 10% to 40%. Overall reduction in CFR bias for MAR across methods, proportion of missingness, outbreak size and proportion of training data was 0.5% (median, range: 0%-11%).<h4>Conclusion</h4>ML methods could reduce bias and increase the precision in CFR estimates at low levels of missingness. However, no method is robust to high percentages of missingness. Thus, a datacentric approach is recommended in outbreak settings-patient survival outcome data should be prioritised for collection and random-sample follow-ups should be implemented to ascertain missing outcomes.Alpha FornaIlaria DorigattiPierre NouvelletChristl A DonnellyPublic Library of Science (PLoS)articleMedicineRScienceQENPLoS ONE, Vol 16, Iss 9, p e0257005 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Alpha Forna Ilaria Dorigatti Pierre Nouvellet Christl A Donnelly Comparison of machine learning methods for estimating case fatality ratios: An Ebola outbreak simulation study.
description	<h4>Background</h4>Machine learning (ML) algorithms are now increasingly used in infectious disease epidemiology. Epidemiologists should understand how ML algorithms behave within the context of outbreak data where missingness of data is almost ubiquitous.<h4>Methods</h4>Using simulated data, we use a ML algorithmic framework to evaluate data imputation performance and the resulting case fatality ratio (CFR) estimates, focusing on the scale and type of data missingness (i.e., missing completely at random-MCAR, missing at random-MAR, or missing not at random-MNAR).<h4>Results</h4>Across ML methods, dataset sizes and proportions of training data used, the area under the receiver operating characteristic curve decreased by 7% (median, range: 1%-16%) when missingness was increased from 10% to 40%. Overall reduction in CFR bias for MAR across methods, proportion of missingness, outbreak size and proportion of training data was 0.5% (median, range: 0%-11%).<h4>Conclusion</h4>ML methods could reduce bias and increase the precision in CFR estimates at low levels of missingness. However, no method is robust to high percentages of missingness. Thus, a datacentric approach is recommended in outbreak settings-patient survival outcome data should be prioritised for collection and random-sample follow-ups should be implemented to ascertain missing outcomes.
format	article
author	Alpha Forna Ilaria Dorigatti Pierre Nouvellet Christl A Donnelly
author_facet	Alpha Forna Ilaria Dorigatti Pierre Nouvellet Christl A Donnelly
author_sort	Alpha Forna
title	Comparison of machine learning methods for estimating case fatality ratios: An Ebola outbreak simulation study.
title_short	Comparison of machine learning methods for estimating case fatality ratios: An Ebola outbreak simulation study.
title_full	Comparison of machine learning methods for estimating case fatality ratios: An Ebola outbreak simulation study.
title_fullStr	Comparison of machine learning methods for estimating case fatality ratios: An Ebola outbreak simulation study.
title_full_unstemmed	Comparison of machine learning methods for estimating case fatality ratios: An Ebola outbreak simulation study.
title_sort	comparison of machine learning methods for estimating case fatality ratios: an ebola outbreak simulation study.
publisher	Public Library of Science (PLoS)
publishDate	2021
url	https://doaj.org/article/f76e19b1b9e84c1795c43352c11e6790
work_keys_str_mv	AT alphaforna comparisonofmachinelearningmethodsforestimatingcasefatalityratiosanebolaoutbreaksimulationstudy AT ilariadorigatti comparisonofmachinelearningmethodsforestimatingcasefatalityratiosanebolaoutbreaksimulationstudy AT pierrenouvellet comparisonofmachinelearningmethodsforestimatingcasefatalityratiosanebolaoutbreaksimulationstudy AT christladonnelly comparisonofmachinelearningmethodsforestimatingcasefatalityratiosanebolaoutbreaksimulationstudy
_version_	1718374682216890368

Comparison of machine learning methods for estimating case fatality ratios: An Ebola outbreak simulation study.

Ejemplares similares