Redes neurais artificiais aplicadas à previsão de surtos de leptospirose

Abstract

Leptospirosis is a disease of global occurrence, especially in tropical regions with high rainfall incidence and in places with poor sanitary conditions. It is a disease considered neglected, underestimated and difficult to diagnose because it presents symptoms common to other diseases, besides having a high lethality rate when reaching serious levels. In Brazil, its treatment has a high hospital cost and the disease has a high incidence in large urban centers. In the state of Pernambuco, Recife, it stands out from the other cities for having significantly larger numbers of records of Leptospirosis infections, this worrying data extols the need to support epidemiological surveillance in the fight against the disease. The objective of this work is to analyze the feasibility of using Artificial Neural Networks applied to the forecast of Time Series to perform early identification of the number of cases of Leptospirosis in the city of Recife / Pernambuco in order to demonstrate the possibility of using the technique for forecasting outbreaks as an additional means of information to contribute to the planning of health prevention actions, minimizing the social and economic impact caused by the disease.