Mapping dengue risk in Singapore using Random Forest

Abstract

BackgroundSingapore experiences endemic dengue, with 2013 being the largest outbreak year known to date, culminating in 22,170 cases. Given the limited resources available, and that vector control is the key approach for prevention in Singapore, it is important that public health professionals know where resources should be invested in. This study aims to stratify the spatial risk of dengue transmission in Singapore for effective deployment of resources.Methodology/principal findingsRandom Forest was used to predict the risk rank of dengue transmission in 1km2 grids, with dengue, population, entomological and environmental data. The predicted risk ranks are categorized and mapped to four color-coded risk groups for easy operation application. The risk maps were evaluated with dengue case and cluster data. Risk maps produced by Random Forest have high accuracy. More than 80% of the observed risk ranks fell within the 80% prediction interval. The observed and predicted risk ranks were highly correlated (≥0.86, P <0.01). Furthermore, the predicted risk levels were in excellent agreement with case density, a weighted Kappa coefficient of more than 0.80 (P <0.01). Close to 90% of the dengue clusters occur in high risk areas, and the odds of cluster forming in high risk areas were higher than in low risk areas.ConclusionsThis study demonstrates the potential of Random Forest and its strong predictive capability in stratifying the spatial risk of dengue transmission in Singapore. Dengue risk map produced using Random Forest has high accuracy, and is a good surveillance tool to guide vector control operations.

Machine learning and dengue forecasting: Comparing random forests and artificial neural networks for predicting dengue burden at national and sub-national scales in Colombia

“…Despite the strengths of our study, our RF approach is likely to generate time lags in forecasting rapid changes in dengue, which is also a common occurrence with other forecasting approaches. Including a predictor of mosquito abundance from an entomological surveillance program may reduce such time lag errors [71]. However, this type of data was not available at the national level given insufficient temporal and spatial granularity.…”

Doenças infecciosas e parasitárias no Brasil de 2010 a 2017: aspectos para vigilância em saúde

“…Para além das limitações, o presente estudo mostrou que a espacialização de doenças de alta magnitude para a realidade brasileira evidencia os contrastes territoriais de condições de saúde associados a fatores determinantes de adoecimentos. A técnica utilizada para identificação de áreas críticas também se mostrou adequada no sentido de orientar as ações de vigilância no âmbito nacional no que tange a necessidade de maior articulação com as vigilâncias locais e demais setores, para implementação de ações conjuntas que possam dirimir os problemas de saúde causados por doenças infecciosas e parasitárias e outros fatores a elas relacionadas (30).…”

Developing a Dengue Forecasting Model: A Case Study in Iligan City

“…A study by [10] used a Random Forest approach that predicts the risk rank of dengue transmission in Singapore with dengue; population, entomological and environmental data. The predicted risk ranks are then categorized and mapped to color-coded risk groups which were evaluated with dengue cases and cluster data.…”

“…Results of the study revealed a 26.9% of the variation of DHF patients using a number of water storage containers, Aedes Aegypti in drainage of refrigerators, pH and temperature of water in container. In the study by Ong et al [7] predicting the Dengue incidence with the use of Random Forest approach, predicted the risk rank of dengue transmission in Singapore with dengue cases, population, entomological and environmental data. The evaluation using the latest dengue case data in the study showed a strong predictive ability for the model, compared to the study results of Tilwani, Dave, & Nadurbarkar [8], that adopted a regression approach with Poisson Regression and Negative Binomial to investigate the correlation between dengue incidence and climatic fluctuations including relative humidity, temperature, and pressure.…”