AI Article Synopsis

  • There is a critical need for scientific analysis to aid governments in managing the COVID-19 pandemic, which this article addresses through a data mining methodology.
  • The study focused on São Paulo, Rio de Janeiro, and Manaus, analyzing environmental factors like air quality and weather that correlate with COVID-19 deaths and cases up to mid-2021.
  • The findings reveal that different environmental conditions can predict COVID-19 outcomes uniquely for each city, and the developed predictive models could be adapted for use in other cities and epidemics.

Article Abstract

There is an ongoing need for scientific analysis to help governments and public health authorities make decisions regarding the COVID-19 pandemic. This article presents a methodology based on data mining that can offer support for coping with epidemic diseases. The methodological approach was applied in São Paulo, Rio de Janeiro and Manaus, the cities in Brazil with the most COVID-19 deaths until the first half of 2021. We aimed to predict the evolution of COVID-19 in metropolises and identify air quality and meteorological variables correlated with confirmed cases and deaths. The statistical analyses indicated the most important explanatory environmental variables, while the cluster analyses showed the potential best input variables for the forecasting models. The forecast models were built by two different algorithms and their results have been compared. The relationship between epidemiological and environmental variables was particular to each of the three cities studied. Low solar radiation periods predicted in Manaus can guide managers to likely increase deaths due to COVID-19. In São Paulo, an increase in the mortality rate can be indicated by drought periods. The developed models can predict new cases and deaths by COVID-19 in studied cities. Furthermore, the methodological approach can be applied in other cities and for other epidemic diseases.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8716530PMC
http://dx.doi.org/10.1038/s41598-021-04029-6DOI Listing

Publication Analysis

Top Keywords

epidemic diseases
8
methodological approach
8
approach applied
8
são paulo
8
cases deaths
8
environmental variables
8
deaths covid-19
8
covid-19
6
data based
4
based model
4

Similar Publications

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!