Application of Machine Learning to predict cases of hospitalization and death due to respiratory diseases in the city of São Paulo, based on the concentration of pollens, pollutants and atmospheric variables - ATIIM.

• Grant number: 24/17102-5
• Support Opportunities: Regular Research Grants
• Start date: April 01, 2025
• End date: March 31, 2028
• Field of knowledge: Physical Sciences and Mathematics - Geosciences
• Principal Investigator: Gregori de Arruda Moreira
• Grantee: Gregori de Arruda Moreira
• Host Institution: Instituto Federal de Educação, Ciência e Tecnologia de São Paulo (IFSP). Campus São Paulo. São Paulo , SP, Brazil
• Associated researchers: Andrés M Vélez Pereira ; Cynthia Fernandes Pinto da Luz ; Eduardo Landulfo ; Juan Luis Guerrero Rascado ; Maria de Fátima Andrade ; MARLON MIGUEL CEDENO PUIG ; Paloma Cariñanos Gonzalez

Abstract

In recent decades, there has been extensive discussion on the relationship between climate change and its impact on human health, especially with the increasing frequency of extreme events. However, there has been limited investigation into the correlation between climate change, pollen concentration, pollutants, and respiratory diseases in Brazil. Despite Brazil's rich biodiversity, there is a significant research gap in understanding how climate change has influenced the main pollen season and the potential effects of different pollen types on the severity of respiratory and allergic diseases. To address this, the project aims to establish two pollen collection stations: one in Parque Estadual das Fontes in Ipiranga and another in the center of São Paulo (Armênia neighborhood). These stations will be equipped with anemometric towers and ceilometers to gather meteorological data. The data collected, the first of its kind in Brazil, will be used to build a comprehensive database, incorporating information on hospitalizations and deaths due to respiratory diseases in São Paulo from DATASuS. Additionally, a Machine Learning model will be developed to predict hospitalizations and deaths and made accessible to the scientific and general community through a web platform. This platform will also offer the collected data to download and disseminate scientific information. (AU)