Assessing the impact of PM2.5 on respiratory disease using artificial neural networks

Gabriela Polezer; Yara S Tadano; Hugo V Siqueira; Ana F L Godoi; Carlos I Yamamoto; Paulo A de André; Theotonio Pauliquevis; Maria de Fatima Andrade; Andrea Oliveira; Paulo H N Saldiva; Philip E Taylor; Ricardo H M Godoi

doi:10.1016/j.envpol.2017.12.111

Assessing the impact of PM_2.5 on respiratory disease using artificial neural networks

Environ Pollut. 2018 Apr:235:394-403. doi: 10.1016/j.envpol.2017.12.111. Epub 2018 Jan 5.

Affiliations

¹ Environmental Engineering Department, Federal University of Parana, 210 Francisco H. dos Santos St., Curitiba, Paraná 81531-980, Brazil.
² Mathematics Department, Federal University of Technology, Ponta Grossa, Paraná, Brazil.
³ Electronic Engineering Department, Federal University of Technology, Ponta Grossa, Paraná, Brazil.
⁴ Chemical Engineering Department, Federal University of Paraná, Curitiba, Paraná, Brazil.
⁵ Department of Pathology, LPAE (Air Pollution Lab), Faculty of Medicine, University of Sao Paulo, São Paulo, Brazil.
⁶ Department of Environmental Sciences, Federal University of Sao Paulo, Diadema, Brazil.
⁷ Department of Atmospheric Sciences, Institute of Astronomy, Geophysics and Atmospheric Sciences, University of São Paulo, São Paulo, Brazil.
⁸ Chemistry Department, Federal University of Parana, Curitiba, Paraná, Brazil.
⁹ Deakin University, School of Life and Environmental Sciences, Geelong, VIC, Australia.
¹⁰ Environmental Engineering Department, Federal University of Parana, 210 Francisco H. dos Santos St., Curitiba, Paraná 81531-980, Brazil. Electronic address: rhmgodoi@ufpr.br.

PMID: 29306807
DOI: 10.1016/j.envpol.2017.12.111

Abstract

Understanding the impact on human health during peak episodes in air pollution is invaluable for policymakers. Particles less than PM_2.5 can penetrate the respiratory system, causing cardiopulmonary and other systemic diseases. Statistical regression models are usually used to assess air pollution impacts on human health. However, when there are databases missing, linear statistical regression may not process well and alternative data processing should be considered. Nonlinear Artificial Neural Networks (ANN) are not employed to research environmental health pollution even though another advantage in using ANN is that the output data can be expressed as the number of hospital admissions. This research applied ANN to assess the impact of air pollution on human health. Three well-known ANN were tested: Multilayer Perceptron (MLP), Extreme Learning Machines (ELM) and Echo State Networks (ESN), to assess the influence of PM_2.5, temperature, and relative humidity on hospital admissions due to respiratory diseases. Daily PM_2.5 levels were monitored, and hospital admissions for respiratory illness were obtained, from the Brazilian hospital information system for all ages during two sampling campaigns (2008-2011 and 2014-2015) in Curitiba, Brazil. During these periods, the daily number of hospital admissions ranged from 2 to 55, PM_2.5 concentrations varied from 0.98 to 54.2 μg m^-3, temperature ranged from 8 to 26 °C, and relative humidity ranged from 45 to 100%. Of the ANN used in this study, MLP gave the best results showing a significant influence of PM_2.5, temperature and humidity on hospital attendance after one day of exposure. The Anova Friedman's test showed statistical difference between the appliance of each ANN model (p < .001) for 1 lag day between PM_2.5 exposure and hospital admission. ANN could be a more sensitive method than statistical regression models for assessing the effects of air pollution on respiratory health, and especially useful when there is limited data available.

Keywords: Artificial neural networks; Health risk assessment; Hospital admissions; PM(2.5); Source identification.

MeSH terms

Air Pollutants / analysis*
Air Pollution / analysis
Air Pollution / statistics & numerical data*
Brazil / epidemiology
Environmental Exposure / statistics & numerical data*
Hospitalization
Humans
Humidity
Linear Models
Models, Statistical
Neural Networks, Computer*
Particulate Matter / analysis*
Regression Analysis
Respiration Disorders / epidemiology
Respiratory Tract Diseases / epidemiology*
Temperature

Substances

Air Pollutants
Particulate Matter