Using artificial intelligence systems to investigate relation between air pollution and acute respiratory symptoms registered at the Emergency Medical Center of Mashhad in 2017

Abstract

Abstract Background and Aim: Air pollution is one of the most significant environmental problems that has a remarkable impact on the incidence of cardiovascular disease and associated mortality. It is essential to comprehend air pollution effects and the ways of emission and predict the number of patients with acute respiratory problems to eliminate and reduce air pollutants and associated mortality. This study aimed to investigate the relationship between different air pollutants and the number of cardiovascular disease patients in Mashhad. Materials and Methods: This study applied a neural network to model and analyze the relationship between CO, NO2, SO2, PM2.5, and PM10 and the number of patients with acute respiratory problems. The inputs were average temperature, humidity, wind direction, and wind speed and the output was the number of people referred per day by gender and age. The data set used included meteorological data from the Iran Meteorological Organization, air pollution data from the Mashhad Meteorological Organization, and the number of daily referrals of heart disease patients to the emergency department of Mashhad. Results: According to this study, the most effective air pollutants in Mashhad were PM2.5 and PM10, followed by NO2, CO, and SO2, respectively. Conclusion: Neural networks can be applied in the modeling of the relationship between environmental parameters and cardiovascular disease patients because they have a high ability to model nonlinear phenomena. These models show that the more airborne particles, the more rate of cardiovascular diseases in Mashhad

Keywords

• air pollution
• acute respiratory diseases
• artificial neural networks
• regression