A Regression Based Approach for Leishmaniasis Outbreak Detection

• Autores: Baptista, Ernie, Vigil, Franco, Ugarte, Willy
• Formato: artículo
• Fecha de Publicación: 2024
• Institución: Universidad Peruana de Ciencias Aplicadas
• Repositorio: UPC-Institucional
• Lenguaje: inglés
• OAI Identifier: oai:repositorioacademico.upc.edu.pe:10757/676005
• Enlace del recurso: http://hdl.handle.net/10757/676005
• Nivel de acceso: acceso embargado
• Materia: Leishmaniasis; Machine Learning; NTDs; Outbreaks; Random Forest

Leishmaniasis is part of a group of diseases called Neglected Tropical Diseases (NTDs) that affects poor and forgotten communities and reports more than 5,000 cases in regions like Brazil, Peru, and Colombia being categorized as endemic in these. In this study, we present a machine-learning model (Random Forest) to predict cases in the future and predict possible outbreaks using meteorological and epidemiological data of the province of la Convencion (Cusco - Peru). Understanding how climate variables affect leishmaniasis outbreaks is an important problem to help people to perform prevention systems. We used several techniques to obtain better metrics and improve our model performance such as synthetic data and hyperparameter optimization. Results showed two important climate factors to analyze and no outbreaks.