Exposure measurement error in air pollution studies: A framework for assessing shared, multiplicative measurement error in ensemble learning estimates of nitrogen oxides.
Journal
Environment international
ISSN: 1873-6750
Titre abrégé: Environ Int
Pays: Netherlands
ID NLM: 7807270
Informations de publication
Date de publication:
04 2019
04 2019
Historique:
received:
07
09
2018
revised:
10
12
2018
accepted:
12
12
2018
pubmed:
4
2
2019
medline:
2
11
2019
entrez:
4
2
2019
Statut:
ppublish
Résumé
Increasingly ensemble learning-based spatiotemporal models are being used to estimate residential air pollution exposures in epidemiological studies. While these machine learning models typically have improved performance, they suffer from exposure measurement error that is inherent in all models. Our objective is to develop a framework to formally assess shared, multiplicative measurement error (SMME) in our previously published three-stage, ensemble learning-based nitrogen oxides (NO By treating the ensembles as an external dosimetry system, we quantified shared and unshared, multiplicative and additive (SUMA) measurement error components in our exposure model. We used generalized additive models (GAMs) with a smooth term for location to identify geographic locations with significantly elevated SMME and explain their spatial and temporal determinants. We found evidence of significant shared and unshared multiplicative error (p < 0.0001) in our ensemble-learning based spatiotemporal NO We developed a novel statistical framework to formally evaluate the magnitude and drivers of SMME in ensemble learning-based exposure models. Our framework can be used to inform building future improved exposure models.
Sections du résumé
BACKGROUND
Increasingly ensemble learning-based spatiotemporal models are being used to estimate residential air pollution exposures in epidemiological studies. While these machine learning models typically have improved performance, they suffer from exposure measurement error that is inherent in all models. Our objective is to develop a framework to formally assess shared, multiplicative measurement error (SMME) in our previously published three-stage, ensemble learning-based nitrogen oxides (NO
METHODS
By treating the ensembles as an external dosimetry system, we quantified shared and unshared, multiplicative and additive (SUMA) measurement error components in our exposure model. We used generalized additive models (GAMs) with a smooth term for location to identify geographic locations with significantly elevated SMME and explain their spatial and temporal determinants.
RESULTS
We found evidence of significant shared and unshared multiplicative error (p < 0.0001) in our ensemble-learning based spatiotemporal NO
CONCLUSIONS
We developed a novel statistical framework to formally evaluate the magnitude and drivers of SMME in ensemble learning-based exposure models. Our framework can be used to inform building future improved exposure models.
Identifiants
pubmed: 30711654
pii: S0160-4120(18)32003-8
doi: 10.1016/j.envint.2018.12.025
pmc: PMC6499078
mid: NIHMS1520458
pii:
doi:
Substances chimiques
Air Pollutants
0
Nitrogen Oxides
0
Types de publication
Journal Article
Research Support, N.I.H., Extramural
Langues
eng
Sous-ensembles de citation
IM
Pagination
97-106Subventions
Organisme : NIEHS NIH HHS
ID : P30 ES007048
Pays : United States
Organisme : NIEHS NIH HHS
ID : P50 ES026086
Pays : United States
Organisme : NIBIB NIH HHS
ID : U54 EB022002
Pays : United States
Organisme : NIH HHS
ID : UH3 OD023287
Pays : United States
Informations de copyright
Copyright © 2019 The Authors. Published by Elsevier Ltd.. All rights reserved.
Références
Environ Res. 2016 Apr;146:1-9
pubmed: 26705853
Air Qual Atmos Health. 2012 Jun;5(2):203-216
pubmed: 22662023
J Allergy Clin Immunol. 2005 Aug;116(2):279-84
pubmed: 16083780
Environ Health Perspect. 2002 Sep;110(9):961-7
pubmed: 12204833
PLoS One. 2015 Mar 23;10(3):e0119418
pubmed: 25799311
Environ Int. 2017 Mar;100:1-31
pubmed: 27881237
Environ Health Perspect. 2017 Apr;125(4):737-745
pubmed: 27219743
Thorax. 2008 Aug;63(8):710-6
pubmed: 18267985
Biometrics. 1997 Sep;53(3):1008-25
pubmed: 9290228
J Thorac Dis. 2015 Jan;7(1):46-58
pubmed: 25694817
Environmetrics. 2009 Sep 1;21(6):606-631
pubmed: 24860253
J Expo Anal Environ Epidemiol. 2004 Sep;14(5):365-77
pubmed: 15361895
Environ Sci Technol. 2017 Sep 5;51(17):9920-9929
pubmed: 28727456
Rev Environ Health. 2000 Jan-Jun;15(1-2):13-42
pubmed: 10939084
Epidemiology. 2008 May;19(3):401-8
pubmed: 18379426
Environ Health Perspect. 2001 Nov;109(11):A517
pubmed: 11762306
Am J Respir Crit Care Med. 2010 Mar 15;181(6):596-603
pubmed: 19965811
Environ Sci Technol. 2017 Jun 20;51(12):6999-7008
pubmed: 28578585
BMC Pulm Med. 2014 Aug 07;14:130
pubmed: 25098250
Environ Sci Technol. 2016 May 3;50(9):4712-21
pubmed: 27023334
Epidemiology. 2015 May;26(3):395-401
pubmed: 25668685
Environ Health Perspect. 2000 May;108(5):419-26
pubmed: 10811568
Rev Environ Health. 1999 Jul-Sep;14(3):121-34
pubmed: 10674285
PLoS One. 2017 Apr 3;12(4):e0174641
pubmed: 28369141
Am J Respir Crit Care Med. 1999 Mar;159(3):760-7
pubmed: 10051248
Radiat Res. 2003 Oct;160(4):408-17
pubmed: 12968933
Biometrics. 2002 Mar;58(1):13-20
pubmed: 11890308
PLoS One. 2018 Feb 6;13(2):e0190792
pubmed: 29408862