Background: Spatiotemporal dengue forecasting using machine learning (ML) can contribute to the development of prevention and control strategies for impending dengue outbreaks. However, training data for dengue incidence may be inflated with frequent zero values because of the rarity of cases, which lowers the prediction accuracy. This study aimed to understand the influence of spatiotemporal resolutions of data on the accuracy of dengue incidence prediction using ML models, to understand how the influence of spatiotemporal resolution differs between quantitative and qualitative predictions of dengue incidence, and to improve the accuracy of dengue incidence prediction with zero-inflated data.
Methodology: We predicted dengue incidence at six spatiotemporal resolutions and compared their prediction accuracy. Six ML algorithms were compared: generalized additive models, random forests, conditional inference forest, artificial neural networks, support vector machines and regression, and extreme gradient boosting. Data from 2009 to 2012 were used for training, and data from 2013 were used for model validation with quantitative and qualitative dengue variables. To address the inaccuracy in the quantitative prediction of dengue incidence due to zero-inflated data at fine spatiotemporal scales, we developed a hybrid approach in which the second-stage quantitative prediction is performed only when/where the first-stage qualitative model predicts the occurrence of dengue cases.
Principal Findings: At higher resolutions, the dengue incidence data were zero-inflated, which was insufficient for quantitative pattern extraction of relationships between dengue incidence and environmental variables by ML. Qualitative models, used as binary variables, eased the effect of data distribution. Our novel hybrid approach of combining qualitative and quantitative predictions demonstrated high potential for predicting zero-inflated or rare phenomena, such as dengue.
Significance: Our research contributes valuable insights to the field of spatiotemporal dengue prediction and provides an alternative solution to enhance prediction accuracy in zero-inflated data where hurdle or zero-inflated models cannot be applied.
Download full-text PDF |
Source |
---|---|
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC11527386 | PMC |
http://dx.doi.org/10.1371/journal.pntd.0012599 | DOI Listing |
Cureus
December 2024
Internal Medicine, Medical Teaching Institution (MTI) Hayatabad Medical Complex, Peshawar, PAK.
Background: Malaria and dengue are significant mosquito-borne diseases prevalent in tropical and subtropical climates, with increasing reports of co-infections. This study aimed to determine the frequency, patterns, and risk factors of these co-infections in Peshawar.
Methods: A cross-sectional study was conducted from June to December 2023 in three tertiary care hospitals in Peshawar.
Viruses
January 2025
Department of Microbiology, Institute of Biomedical Sciences, University of São Paulo, São Paulo 5508-900, Brazil.
Dengue fever, caused by the dengue virus (DENV), poses a significant global health challenge, particularly in tropical and subtropical regions. Recent increases in indigenous DENV cases in Europe are concerning, reflecting rising incidence linked to climate change and the spread of mosquitoes. These vectors thrive under environmental conditions like temperature and humidity, which are increasingly influenced by climate change.
View Article and Find Full Text PDFViruses
December 2024
Department of Epidemiology, Gillings School of Global Public Health, The University of North Carolina at Chapel Hill, Chapel Hill, NC 27599-3500, USA.
Flaviviruses are arthropod-borne viruses primarily transmitted through the mosquito or genus of mosquitos. These viruses are predominantly found in tropical and subtropical regions of the world with their geographical spread predicted to increase as global temperatures continue to rise. These viruses cause a variety of diseases in humans with the most prevalent being caused by dengue, resulting in hemorrhagic fever and associated sequala.
View Article and Find Full Text PDFPathogens
January 2025
Department of Clinical and Experimental Sciences, Unit of Infectious and Tropical Diseases, University of Brescia, ASST Spedali Civili di Brescia, 25123 Brescia, Italy.
The rise and resurgence of vector-borne diseases (VBDs) in Europe pose an expanding public health challenge, exacerbated by climate change, globalization, and ecological disruptions. Both arthropod-borne viruses (arboviruses) transmitted by ticks such as Crimean-Congo hemorrhagic fever and arboviruses transmitted by mosquitoes like dengue, Chikungunya, Zika, and Japanese encephalitis have broadened their distribution due to rising temperatures, changes in rainfall, and increased human mobility. By emphasizing the importance of interconnected human, animal, and environmental health, integrated One Health strategies are crucial in addressing this complex issue.
View Article and Find Full Text PDFPathogens
January 2025
Health and Biosecurity, Commonwealth Scientific and Industrial Research Organisation, Geelong, VIC 3220, Australia.
Current arbovirus surveillance strategies in Australia involve mosquito collection, species identification, and virus detection. These processes are labour-intensive, expensive, and time-consuming and can lead to delays in reporting. Mosquito excreta has been proposed as an alternative sample type to whole mosquito collection, with potential to streamline the virus surveillance pipeline.
View Article and Find Full Text PDFEnter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!