Dataset Information

Deep learning models for forecasting dengue fever based on climate data in Vietnam.

ABSTRACT:

Background

Dengue fever (DF) represents a significant health burden in Vietnam, which is forecast to worsen under climate change. The development of an early-warning system for DF has been selected as a prioritised health adaptation measure to climate change in Vietnam.

Objective

This study aimed to develop an accurate DF prediction model in Vietnam using a wide range of meteorological factors as inputs to inform public health responses for outbreak prevention in the context of future climate change.

Methods

Convolutional neural network (CNN), Transformer, long short-term memory (LSTM), and attention-enhanced LSTM (LSTM-ATT) models were compared with traditional machine learning models on weather-based DF forecasting. Models were developed using lagged DF incidence and meteorological variables (measures of temperature, humidity, rainfall, evaporation, and sunshine hours) as inputs for 20 provinces throughout Vietnam. Data from 1997-2013 were used to train models, which were then evaluated using data from 2014-2016 by Root Mean Square Error (RMSE) and Mean Absolute Error (MAE).

Results and discussion

LSTM-ATT displayed the highest performance, scoring average places of 1.60 for RMSE-based ranking and 1.95 for MAE-based ranking. Notably, it was able to forecast DF incidence better than LSTM in 13 or 14 out of 20 provinces for MAE or RMSE, respectively. Moreover, LSTM-ATT was able to accurately predict DF incidence and outbreak months up to 3 months ahead, though performance dropped slightly compared to short-term forecasts. To the best of our knowledge, this is the first time deep learning methods have been employed for the prediction of both long- and short-term DF incidence and outbreaks in Vietnam using unique, rich meteorological features.

Conclusion

This study demonstrates the usefulness of deep learning models for meteorological factor-based DF forecasting. LSTM-ATT should be further explored for mitigation strategies against DF and other climate-sensitive diseases in the coming years.

SUBMITTER: Nguyen VH

PROVIDER: S-EPMC9232166 | biostudies-literature | 2022 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Deep learning models for forecasting dengue fever based on climate data in Vietnam.

Nguyen Van-Hau VH Tuyet-Hanh Tran Thi TT Mulhall James J Mulhall James J Minh Hoang Van HV Duong Trung Q TQ Chien Nguyen Van NV Nhung Nguyen Thi Trang NTT Lan Vu Hoang VH Minh Hoang Ba HB Cuong Do D Bich Nguyen Ngoc NN Quyen Nguyen Huu NH Linh Tran Nu Quy TNQ Tho Nguyen Thi NT Nghia Ngu Duy ND Anh Le Van Quoc LVQ Phan Diep T M DTM Hung Nguyen Quoc Viet NQV Son Mai Thai MT

PLoS neglected tropical diseases 20220613 6

<h4>Background</h4>Dengue fever (DF) represents a significant health burden in Vietnam, which is forecast to worsen under climate change. The development of an early-warning system for DF has been selected as a prioritised health adaptation measure to climate change in Vietnam.<h4>Objective</h4>This study aimed to develop an accurate DF prediction model in Vietnam using a wide range of meteorological factors as inputs to inform public health responses for outbreak prevention in the context of fu ...[more]

PMID: 35696432

Similar Datasets

Project description:BackgroundDengue dynamics are driven by complex interactions between human-hosts, mosquito-vectors and viruses that are influenced by environmental and climatic factors. The objectives of this study were to analyze and model the relationships between climate, Aedes aegypti vectors and dengue outbreaks in Noumea (New Caledonia), and to provide an early warning system.Methodology/principal findingsEpidemiological and meteorological data were analyzed from 1971 to 2010 in Noumea. Entomological surveillance indices were available from March 2000 to December 2009. During epidemic years, the distribution of dengue cases was highly seasonal. The epidemic peak (March-April) lagged the warmest temperature by 1-2 months and was in phase with maximum precipitations, relative humidity and entomological indices. Significant inter-annual correlations were observed between the risk of outbreak and summertime temperature, precipitations or relative humidity but not ENSO. Climate-based multivariate non-linear models were developed to estimate the yearly risk of dengue outbreak in Noumea. The best explicative meteorological variables were the number of days with maximal temperature exceeding 32°C during January-February-March and the number of days with maximal relative humidity exceeding 95% during January. The best predictive variables were the maximal temperature in December and maximal relative humidity during October-November-December of the previous year. For a probability of dengue outbreak above 65% in leave-one-out cross validation, the explicative model predicted 94% of the epidemic years and 79% of the non epidemic years, and the predictive model 79% and 65%, respectively.Conclusions/significanceThe epidemic dynamics of dengue in Noumea were essentially driven by climate during the last forty years. Specific conditions based on maximal temperature and relative humidity thresholds were determinant in outbreaks occurrence. Their persistence was also crucial. An operational model that will enable health authorities to anticipate the outbreak risk was successfully developed. Similar models may be developed to improve dengue management in other countries.

Project description:BackgroundWith enough advanced notice, dengue outbreaks can be mitigated. As a climate-sensitive disease, environmental conditions and past patterns of dengue can be used to make predictions about future outbreak risk. These predictions improve public health planning and decision-making to ultimately reduce the burden of disease. Past approaches to dengue forecasting have used seasonal climate forecasts, but the predictive ability of a system using different lead times in a year-round prediction system has been seldom explored. Moreover, the transition from theoretical to operational systems integrated with disease control activities is rare.Methods and findingsWe introduce an operational seasonal dengue forecasting system for Vietnam where Earth observations, seasonal climate forecasts, and lagged dengue cases are used to drive a superensemble of probabilistic dengue models to predict dengue risk up to 6 months ahead. Bayesian spatiotemporal models were fit to 19 years (2002-2020) of dengue data at the province level across Vietnam. A superensemble of these models then makes probabilistic predictions of dengue incidence at various future time points aligned with key Vietnamese decision and planning deadlines. We demonstrate that the superensemble generates more accurate predictions of dengue incidence than the individual models it incorporates across a suite of time horizons and transmission settings. Using historical data, the superensemble made slightly more accurate predictions (continuous rank probability score [CRPS] = 66.8, 95% CI 60.6-148.0) than a baseline model which forecasts the same incidence rate every month (CRPS = 79.4, 95% CI 78.5-80.5) at lead times of 1 to 3 months, albeit with larger uncertainty. The outbreak detection capability of the superensemble was considerably larger (69%) than that of the baseline model (54.5%). Predictions were most accurate in southern Vietnam, an area that experiences semi-regular seasonal dengue transmission. The system also demonstrated added value across multiple areas compared to previous practice of not using a forecast. We use the system to make a prospective prediction for dengue incidence in Vietnam for the period May to October 2020. Prospective predictions made with the superensemble were slightly more accurate (CRPS = 110, 95% CI 102-575) than those made with the baseline model (CRPS = 125, 95% CI 120-168) but had larger uncertainty. Finally, we propose a framework for the evaluation of probabilistic predictions. Despite the demonstrated value of our forecasting system, the approach is limited by the consistency of the dengue case data, as well as the lack of publicly available, continuous, and long-term data sets on mosquito control efforts and serotype-specific case data.ConclusionsThis study shows that by combining detailed Earth observation data, seasonal climate forecasts, and state-of-the-art models, dengue outbreaks can be predicted across a broad range of settings, with enough lead time to meaningfully inform dengue control. While our system omits some important variables not currently available at a subnational scale, the majority of past outbreaks could be predicted up to 3 months ahead. Over the next 2 years, the system will be prospectively evaluated and, if successful, potentially extended to other areas and other climate-sensitive disease systems.

Project description:BACKGROUND:Dengue fever is the most widespread infectious disease of humans transmitted by Aedes mosquitoes. It is the leading cause of hospitalization and death in children in the Southeast Asia and western Pacific regions. We analyzed surveillance records from health centers in Vietnam collected between 2001-2012 to determine seasonal trends, develop risk maps and an incidence forecasting model. METHODS:The data were analyzed using a hierarchical spatial Bayesian model that approximates its posterior parameter distributions using the integrated Laplace approximation algorithm (INLA). Meteorological, altitude and land cover (LC) data were used as predictors. The data were grouped by province (n = 63) and month (n = 144) and divided into training (2001-2009) and validation (2010-2012) sets. Thirteen meteorological variables, 7 land cover data and altitude were considered as predictors. Only significant predictors were kept in the final multivariable model. Eleven dummy variables representing month were also fitted to account for seasonal effects. Spatial and temporal effects were accounted for using Besag-York-Mollie (BYM) and autoregressive (1) models. Their levels of significance were analyzed using deviance information criterion (DIC). The model was validated based on the Theil's coefficient which compared predicted and observed incidence estimated using the validation data. Dengue incidence predictions for 2010-2012 were also used to generate risk maps. RESULTS:The mean monthly dengue incidence during the period was 6.94 cases (SD 14.49) per 100,000 people. Analyses on the temporal trends of the disease showed regular seasonal epidemics that were interrupted every 3 years (specifically in July 2004, July 2007 and September 2010) by major fluctuations in incidence. Monthly mean minimum temperature, rainfall, area under urban settlement/build-up areas and altitude were significant in the final model. Minimum temperature and rainfall had non-linear effects and lagging them by two months provided a better fitting model compared to using unlagged variables. Forecasts for the validation period closely mirrored the observed data and accurately captured the troughs and peaks of dengue incidence trajectories. A favorable Theil's coefficient of inequality of 0.22 was generated. CONCLUSIONS:The study identified temperature, rainfall, altitude and area under urban settlement as being significant predictors of dengue incidence. The statistical model fitted the data well based on Theil's coefficient of inequality, and risk maps generated from its predictions identified most of the high-risk provinces throughout the country.

Dataset Information

Deep learning models for forecasting dengue fever based on climate data in Vietnam.

Background

Objective

Methods

Results and discussion

Conclusion

Publications

Deep learning models for forecasting dengue fever based on climate data in Vietnam.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets