Machine learning methods reveal the temporal pattern of dengue incidence using meteorological factors in metropolitan Manila, Philippines

Effects of El Niño and the Positive Indian Ocean Dipole (+IOD) on Health, Food Security, Economics, and Conflict in Low- and Middle-Income Countries in the Indo-Pacific: A Systematic Review

Climate drivers such as the El Niño Southern Oscillation (ENSO) and the Indian Ocean Dipole (IOD) can impact multiple sectors globally. We are currently witnessing the effects of these teleconnections against the backdrop of a changing climate. This systematic review takes stock of the available evidence on compounding and cascading effects of El Niño and the Positive IOD on health, economic, migration, conflicts, and nutrition outcomes in low- and middle-income countries from the Indo-Pacific region. The review sheds light on how effects vary between and within the considered countries and explores potential sources of heterogeneity. The search of studies was carried out in January 2024 in 12 major databases/search engines and 14 institutional websites, using English keywords, and paired by forward and backward citation tracking of the included studies. The review's inclusion criteria encompassed quantitative studies as long as they provide an estimate of relationship between the climate driver and outcome, and qualitative studies that aim to infer causation such as realist evaluation or process tracing. The analysis used a combination of meta-analysis with random-effects models, median effects from correlational and regression studies, and narrative synthesis. We found that El Niño is likely to decrease agricultural production and productivity at the Indo-Pacific level, although the analysed studies are highly diverse. The absence of evidence on the effects of the considered climate drivers on migration, conflict, food security and nutrition is an important evidence gap. We found limited evidence on the differential effects by El Niño's and +IOD's magnitude and no studies examining their combined impact or qualitative effectiveness studies. The high risk of bias detected across studies calls for more thorough attention to study design, conduct, and reporting in answering questions about effects. Despite remaining evidence gaps, this review highlights potential effects of El Niño and +IOD in the Indo-Pacific and underscores the need for context-specific policy responses to mitigate risks at local and regional levels. Caution is warranted in interpreting the overall findings given the generally high risk of bias of evidence.

Forecasting invasive mosquito abundance in the Basque Country, Spain using machine learning techniques

BACKGROUND

Mosquito-borne diseases cause millions of deaths each year and are increasingly spreading from tropical and subtropical regions into temperate zones, posing significant public health risks. In the Basque Country region of Spain, changing climatic conditions have driven the spread of invasive mosquitoes, increasing the potential for local transmission of diseases such as dengue, Zika, and chikungunya. The establishment of mosquito species in new areas, coupled with rising mosquito populations and viremic imported cases, presents challenges for public health systems in non-endemic regions.

METHODS

This study uses models that capture the complexities of the mosquito life cycle, driven by interactions with weather variables, including temperature, precipitation, and humidity. Leveraging machine learning techniques, we aimed to forecast Aedes invasive mosquito abundance in the provinces of the Basque Country, using egg count as a proxy and weather features as key independent variables. A Spearman correlation was used to assess relationships between climate variables and mosquito egg counts, as well as their lagged time series versions. Forecasting models, including random forest (RF) and seasonal autoregressive integrated moving average (SARIMAX), were evaluated using root mean squared error (RMSE) and mean absolute error (MAE) metrics.

RESULTS

Statistical analysis revealed significant impacts of temperature, precipitation, and humidity on mosquito egg abundance. The random forest (RF) model demonstrated the highest forecasting accuracy, followed by the SARIMAX model. Incorporating lagged climate variables and ovitrap egg counts into the models improved predictions, enabling more accurate forecasts of Aedes invasive mosquito abundance.

CONCLUSIONS

The findings emphasize the importance of integrating climate-driven forecasting tools to predict the abundance of mosquitoes where data are available. Furthermore, this study highlights the critical need for ongoing entomological surveillance to enhance mosquito spread forecasting and contribute to the development and assessment of effective vector control strategies in regions of mosquito expansion.

Hybrid Machine Learning Approach to Zero-Inflated Data Improves Accuracy of Dengue Prediction

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.

Spatiotemporal models of dengue epidemiology in the Philippines: Integrating remote sensing and interpretable machine learning

Previous dengue epidemiological analyses have been limited in spatiotemporal extent or covariate dimensions, the latter neglecting the multifactorial nature of dengue. These constraints, caused by rigid and traditional statistical tools which collapse amidst 'Big Data', prompt interpretable machine-learning (iML) approaches. Predicting dengue incidence and mortality in the Philippines, a data-limited yet high-burden country, the mlr3 universe of R packages was used to build and optimize ML models based on remotely sensed provincial and dekadal 3 NDVI and 9 rainfall features from 2016 to 2020. Between two tasks, models differ across four random forest-based learners and two clustering strategies. Among 16 candidates, rfsrc-year-case and ranger-year-death significantly perform best for predicting dengue incidence and mortality, respectively. Therefore, temporal clustering yields the best models, reflective of dengue seasonality. The two best models were subjected to tripartite global exploratory model analyses, which encompass model-agnostic post-hoc methods such as Permutation Feature Importance (PFI) and Accumulated Local Effects (ALE). PFI reveals that the models differ in their important explanatory aspect, rainfall for rfsrc-year-case and NDVI for ranger-year-death, among which long-term average (lta) features are most relevant. Trend-wise, ALE reveals that average incidence predictions are positively associated with 'Rain.lta', reflective of dengue cases peaking during the wet season. In contrast, those for mortality are negatively associated with 'NDVI.lta', reflective of urban spaces driving dengue-related deaths. By technologically addressing the challenges of the human-animal-ecosystem interface, this study adheres to the One Digital Health paradigm operationalized under Sustainable Development Goals (SDGs). Leveraging data digitization and predictive modeling for epidemiological research paves SDG 3, which prioritizes holistic health and well-being.

Improving dengue fever predictions in Taiwan based on feature selection and random forests

BACKGROUND

Dengue fever is a well-studied vector-borne disease in tropical and subtropical areas of the world. Several methods for predicting the occurrence of dengue fever in Taiwan have been proposed. However, to the best of our knowledge, no study has investigated the relationship between air quality indices (AQIs) and dengue fever in Taiwan.

RESULTS

This study aimed to develop a dengue fever prediction model in which meteorological factors, a vector index, and AQIs were incorporated into different machine learning algorithms. A total of 805 meteorological records from 2013 to 2015 were collected from government open-source data after preprocessing. In addition to well-known dengue-related factors, we investigated the effects of novel variables, including particulate matter with an aerodynamic diameter < 10 µm (PM10), PM2.5, and an ultraviolet index, for predicting dengue fever occurrence. The collected dataset was randomly divided into an 80% training set and a 20% test set. The experimental results showed that the random forests achieved an area under the receiver operating characteristic curve of 0.9547 for the test set, which was the best compared with the other machine learning algorithms. In addition, the temperature was the most important factor in our variable importance analysis, and it showed a positive effect on dengue fever at < 30 °C but had less of an effect at > 30 °C. The AQIs were not as important as temperature, but one was selected in the process of filtering the variables and showed a certain influence on the final results.

CONCLUSIONS

Our study is the first to demonstrate that AQI negatively affects dengue fever occurrence in Taiwan. The proposed prediction model can be used as an early warning system for public health to prevent dengue fever outbreaks.

Meteorological factors cannot be ignored in machine learning-based methods for predicting dengue, a systematic review

In recent years, there has been a rapid increase in the application of machine learning methods about predicting the incidence of dengue fever. However, the predictive factors and models employed in different studies vary greatly. Hence, we conducted a systematic review to summarize machine learning methods and predictors in previous studies. We searched PubMed, ScienceDirect, and Web of Science databases for articles published up to July 2023. The selected papers included not only the forecast of dengue incidence but also machine learning methods. A total of 23 papers were included in this study. Predictive factors included meteorological factors (22, 95.7%), historical dengue data (14, 60.9%), environmental factors (4, 17.4%), socioeconomic factors (4, 17.4%), vector surveillance data (2, 8.7%), and internet search data (3, 13.0%). Among meteorological factors, temperature (20, 87.0%), rainfall (20, 87.0%), and relative humidity (14, 60.9%) were the most commonly used. We found that Support Vector Machine (SVM) (6, 26.1%), Long Short-Term Memory (LSTM) (5, 21.7%), Random Forest (RF) (4, 17.4%), Least Absolute Shrinkage and Selection Operator (LASSO) (2, 8.7%), ensemble model (2, 8.7%), and other models (4, 17.4%) were identified as the best models based on evaluation metrics used in each article. These results indicate that meteorological factors are important predictors that cannot be ignored and SVM and LSTM algorithms are the most commonly used models in dengue fever prediction with good predictive performance. This review will contribute to the development of more robust early dengue warning systems and promote the application of machine learning methods in predicting climate-related infectious diseases.

Modeling the Role of Weather and Pilgrimage Variables on Dengue Fever Incidence in Saudi Arabia

The first case of dengue fever (DF) in Saudi Arabia appeared in 1993 but by 2022, DF incidence was 11 per 100,000 people. Climatologic and population factors, such as the annual Hajj, likely contribute to DF's epidemiology in Saudi Arabia. In this study, we assess the impact of these variables on the DF burden of disease in Saudi Arabia and we attempt to create robust DF predictive models. Using 10 years of DF, weather, and pilgrimage data, we conducted a bivariate analysis investigating the role of weather and pilgrimage variables on DF incidence. We also compared the abilities of three different predictive models. Amongst weather variables, temperature and humidity had the strongest associations with DF incidence, while rainfall showed little to no significant relationship. Pilgrimage variables did not have strong associations with DF incidence. The random forest model had the highest predictive ability (R2 = 0.62) when previous DF data were withheld, and the ARIMA model was the best (R2 = 0.78) when previous DF data were incorporated. We found that a nonlinear machine-learning model incorporating temperature and humidity variables had the best prediction accuracy for DF, regardless of the availability of previous DF data. This finding can inform DF early warning systems and preparedness in Saudi Arabia.

The impact of anthropogenic climate change on pediatric viral diseases

The adverse effects of climate change on human health are unfolding in real time. Environmental fragmentation is amplifying spillover of viruses from wildlife to humans. Increasing temperatures are expanding mosquito and tick habitats, introducing vector-borne viruses into immunologically susceptible populations. More frequent flooding is spreading water-borne viral pathogens, while prolonged droughts reduce regional capacity to prevent and respond to disease outbreaks with adequate water, sanitation, and hygiene resources. Worsening air quality and altered transmission seasons due to an increasingly volatile climate may exacerbate the impacts of respiratory viruses. Furthermore, both extreme weather events and long-term climate variation are causing the destruction of health systems and large-scale migrations, reshaping health care delivery in the face of an evolving global burden of viral disease. Because of their immunological immaturity, differences in physiology (e.g., size), dependence on caregivers, and behavioral traits, children are particularly vulnerable to climate change. This investigation into the unique pediatric viral threats posed by an increasingly inhospitable world elucidates potential avenues of targeted programming and uncovers future research questions to effect equitable, actionable change. IMPACT: A review of the effects of climate change on viral threats to pediatric health, including zoonotic, vector-borne, water-borne, and respiratory viruses, as well as distal threats related to climate-induced migration and health systems. A unique focus on viruses offers a more in-depth look at the effect of climate change on vector competence, viral particle survival, co-morbidities, and host behavior. An examination of children as a particularly vulnerable population provokes programming tailored to their unique set of vulnerabilities and encourages reflection on equitable climate adaptation frameworks.

Scalable Lightweight IoT-Based Smart Weather Measurement System

The Internet of Things (IoT) plays a critical role in remotely monitoring a wide variety of different application sectors, including agriculture, building, and energy. The wind turbine energy generator (WTEG) is a real-world application that can take advantage of IoT technologies, such as a low-cost weather station, where human activities can be significantly affected by enhancing the production of clean energy based on the known direction of the wind. Meanwhile, common weather stations are neither affordable nor customizable for specific applications. Moreover, due to weather forecast changes over time and location within the same city, it is not efficient to rely on a limited number of weather stations that may be located far away from a recipient's location. Therefore, in this paper, we focus on presenting a low-cost weather station that relies on an artificial intelligence (AI) algorithm that can be distributed across a WTEG area with minimal cost. The proposed study measures multiple weather parameters, such as wind direction, wind velocity (WV), temperature, pressure, mean sea level, and relative humidity to provide current measurements to recipients and AI-based forecasts. In addition, the proposed study consists of several heterogeneous nodes and a controller for each station in a target area. The collected data can be transmitted through Bluetooth low energy (BLE). The experimental results reveal that the proposed study matches the standard of the National Meteorological Center (NMC), with a nowcast measurement of 95% accuracy for WV and 92% for wind direction (WD).

DDPM: A Dengue Disease Prediction and Diagnosis Model Using Sentiment Analysis and Machine Learning Algorithms

The aedes mosquito-borne dengue viruses cause dengue fever, an arboviral disease (DENVs). In 2019, the World Health Organization forecasts a yearly occurrence of infections from 100 million to 400 million, the maximum number of dengue cases ever testified worldwide, prompting WHO to label the virus one of the world’s top ten public health risks. Dengue hemorrhagic fever can progress into dengue shock syndrome, which can be fatal. Dengue hemorrhagic fever can also advance into dengue shock syndrome. To provide accessible and timely supportive care and therapy, it is necessary to have indispensable practical instruments that accurately differentiate Dengue and its subcategories in the early stages of illness development. Dengue fever can be predicted in advance, saving one’s life by warning them to seek proper diagnosis and treatment. Predicting infectious diseases such as dengue is difficult, and most forecast systems are still in their primary stages. In developing dengue predictive models, data from microarrays and RNA-Seq have been used significantly. Bayesian inferences and support vector machine algorithms are two examples of statistical methods that can mine opinions and analyze sentiment from text. In general, these methods are not very strong semantically, and they only work effectively when the text passage inputs are at the level of the page or the paragraph; they are poor miners of sentiment at the level of the sentence or the phrase. In this research, we propose to construct a machine learning method to forecast dengue fever.

Leptospirosis modelling using hydrometeorological indices and random forest machine learning

Leptospirosis is a zoonosis that has been linked to hydrometeorological variability. Hydrometeorological averages and extremes have been used before as drivers in the statistical prediction of disease. However, their importance and predictive capacity are still little known. In this study, the use of a random forest classifier was explored to analyze the relative importance of hydrometeorological indices in developing the leptospirosis model and to evaluate the performance of models based on the type of indices used, using case data from three districts in Kelantan, Malaysia, that experience annual monsoonal rainfall and flooding. First, hydrometeorological data including rainfall, streamflow, water level, relative humidity, and temperature were transformed into 164 weekly average and extreme indices in accordance with the Expert Team on Climate Change Detection and Indices (ETCCDI). Then, weekly case occurrences were classified into binary classes "high" and "low" based on an average threshold. Seventeen models based on "average," "extreme," and "mixed" indices were trained by optimizing the feature subsets based on the model computed mean decrease Gini (MDG) scores. The variable importance was assessed through cross-correlation analysis and the MDG score. The average and extreme models showed similar prediction accuracy ranges (61.5-76.1% and 72.3-77.0%) while the mixed models showed an improvement (71.7-82.6% prediction accuracy). An extreme model was the most sensitive while an average model was the most specific. The time lag associated with the driving indices agreed with the seasonality of the monsoon. The rainfall variable (extreme) was the most important in classifying the leptospirosis occurrence while streamflow was the least important despite showing higher correlations with leptospirosis.

A systematic review of dengue outbreak prediction models: Current scenario and future directions

Dengue is among the fastest-spreading vector-borne infectious disease, with outbreaks often overwhelm the health system and result in huge morbidity and mortality in its endemic populations in the absence of an efficient warning system. A large number of prediction models are currently in use globally. As such, this study aimed to systematically review the published literature that used quantitative models to predict dengue outbreaks and provide insights about the current practices. A systematic search was undertaken, using the Ovid MEDLINE, EMBASE, Scopus and Web of Science databases for published citations, without time or geographical restrictions. Study selection, data extraction and management process were devised in accordance with the 'Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies' ('CHARMS') framework. A total of 99 models were included in the review from 64 studies. Most models sourced climate (94.7%) and climate change (77.8%) data from agency reports and only 59.6% of the models adjusted for reporting time lag. All included models used climate predictors; 70.7% of them were built with only climate factors. Climate factors were used in combination with climate change factors (13.4%), both climate change and demographic factors (3.1%), vector factors (6.3%), and demographic factors (5.2%). Machine learning techniques were used for 39.4% of the models. Of these, random forest (15.4%), neural networks (23.1%) and ensemble models (10.3%) were notable. Among the statistical (60.6%) models, linear regression (18.3%), Poisson regression (18.3%), generalized additive models (16.7%) and time series/autoregressive models (26.7%) were notable. Around 20.2% of the models reported no validation at all and only 5.2% reported external validation. The reporting of methodology and model performance measures were inadequate in many of the existing prediction models. This review collates plausible predictors and methodological approaches, which will contribute to robust modelling in diverse settings and populations.

Weather integrated multiple machine learning models for prediction of dengue prevalence in India

Dengue is a rapidly spreading viral disease transmitted to humans by Aedes mosquitoes. Due to global urbanization and climate change, the number of dengue cases are gradually increasing in recent decades. Hence, an early prediction of dengue continues to be a major concern for public health in countries with high prevalence of dengue. Creating a robust forecast model for the accurate prediction of dengue is a complex task and can be done through various data modelling approaches. In the present study, we have applied vector auto regression, generalized boosted models, support vector regression, and long short-term memory (LSTM) to predict the dengue prevalence in Kerala state of the Indian subcontinent. We consider the number of dengue cases as the target variable and weather variables viz., relative humidity, soil moisture, mean temperature, precipitation, and NINO3.4 as independent variables. Various analytical models have been applied on both datasets and predicted the dengue cases. Among all the models, the LSTM model was outperformed with superior prediction capability (RMSE: 0.345 and R²:0.86) than the other models. However, other models are able to capture the trend of dengue cases but failed in predicting the outbreak periods when compared to LSTM. The findings of this study will be helpful for public health agencies and policymakers to draw appropriate control measures before the onset of dengue. The proposed LSTM model for dengue prediction can be followed by other states of India as well.

Nonlinear Effects of the Neighborhood Environments on Residents' Mental Health

In the context of rapid urbanization and the "Healthy China" strategy, neighborhood environments play an important role in improving mental health among urban residents. While an increasing number of studies have explored the linear relationships between neighborhood environments and mental health, much remains to be revealed about the nonlinear health effects of neighborhood environments, the thresholds of various environmental factors, and the optimal environmental exposure levels for residents. To fill these gaps, this paper collected survey data from 1003 adult residents in Guangzhou, China, and measured the built and social environments within the neighborhoods. The random forest model was then employed to examine the nonlinear effects of neighborhood environments on mental health, evaluate the importance of each environmental variable, as well as identify the thresholds and optimal levels of various environmental factors. The results indicated that there are differences in the importance of diverse neighborhood environmental factors affecting mental health, and the more critical environmental factors included greenness, neighborhood communication, and fitness facility density. The nonlinear effects were shown to be universal and varied among neighborhood environmental factors, which could be classified into two categories: (i) higher exposure levels of some environmental factors (e.g., greenness, neighborhood communication, and neighborhood safety) were associated with better mental health; (ii) appropriate exposure levels of some environmental factors (e.g., medical, fitness, and entertainment facilities, and public transport stations) had positive effects on mental health, whereas a much higher or lower exposure level exerted a negative impact. Additionally, this study identified the exact thresholds and optimal exposure levels of neighborhood environmental factors, such as the threshold (22.00%) and optimal exposure level (>22.00%) of greenness and the threshold (3.80 number/km²) and optimal exposure level (3.80 number/km²) of fitness facility density.

Forecasting Weekly Dengue Cases by Integrating Google Earth Engine-Based Risk Predictor Generation and Google Colab-Based Deep Learning Modeling in Fortaleza and the Federal District, Brazil

Efficient and accurate dengue risk prediction is an important basis for dengue prevention and control, which faces challenges, such as downloading and processing multi-source data to generate risk predictors and consuming significant time and computational resources to train and validate models locally. In this context, this study proposed a framework for dengue risk prediction by integrating big geospatial data cloud computing based on Google Earth Engine (GEE) platform and artificial intelligence modeling on the Google Colab platform. It enables defining the epidemiological calendar, delineating the predominant area of dengue transmission in cities, generating the data of risk predictors, and defining multi-date ahead prediction scenarios. We implemented the experiments based on weekly dengue cases during 2013-2020 in the Federal District and Fortaleza, Brazil to evaluate the performance of the proposed framework. Four predictors were considered, including total rainfall (R_sum), mean temperature (T_mean), mean relative humidity (RH_mean), and mean normalized difference vegetation index (NDVI_mean). Three models (i.e., random forest (RF), long-short term memory (LSTM), and LSTM with attention mechanism (LSTM-ATT)), and two modeling scenarios (i.e., modeling with or without dengue cases) were set to implement 1- to 4-week ahead predictions. A total of 24 models were built, and the results showed in general that LSTM and LSTM-ATT models outperformed RF models; modeling could benefit from using historical dengue cases as one of the predictors, and it makes the predicted curve fluctuation more stable compared with that only using climate and environmental factors; attention mechanism could further improve the performance of LSTM models. This study provides implications for future dengue risk prediction in terms of the effectiveness of GEE-based big geospatial data processing for risk predictor generation and Google Colab-based risk modeling and presents the benefits of using historical dengue data as one of the input features and the attention mechanism for LSTM modeling.

The practicality of Malaysia dengue outbreak forecasting model as an early warning system

Dengue is a harmful tropical disease that causes death to many people. Currently, the dengue vaccine development is still at an early stage, and only intervention methods exist after dengue cases increase. Thus, previously, two scientific experimental field studies were conducted in producing a dengue outbreak forecasting model as an early warning system. Successfully, an Autoregressive Distributed Lag (ADL) Model was developed using three factors: the epidemiological, entomological, and environmental with an accuracy of 85%; but a higher percentage is required in minimizing the error for the model to be useful. Hence, this study aimed to develop a practical and cost-effective dengue outbreak forecasting model with at least 90% accuracy to be embedded in an early warning computer system using the Internet of Things (IoT) approach. Eighty-one weeks of time series data of the three factors were used in six forecasting models, which were Autoregressive Distributed Lag (ADL), Hierarchical Forecasting (Bottom-up and Optimal combination) and three Machine Learning methods: (Artificial Neural Network (ANN), Support Vector Machine (SVM) and Random Forest). Five error measures were used to evaluate the consistency performance of the models in order to ensure model performance. The findings indicated Random Forest outperformed the other models with an accuracy of 95% when including all three factors. But practically, collecting mosquito related data (the entomological factor) was very costly and time consuming. Thus, it was removed from the model, and the accuracy dropped to 92% but still high enough to be of practical use, i.e., beyond 90%. However, the practical ground operationalization of the early warning system also requires several rain gauges to be located at the dengue hot spots due to localized rainfall. Hence, further analysis was conducted in determining the location of the rain gauges. This has led to the recommendation that the rain gauges should be located about 3–4 km apart at the dengue hot spots to ensure the accuracy of the rainfall data to be included in the dengue outbreak forecasting model so that it can be embedded in the early warning system. Therefore, this early warning system can save lives, and prevention is better than cure.

Climate Change and Cascading Risks from Infectious Disease

Climate change is adversely affecting the burden of infectious disease throughout the world, which is a health security threat. Climate-sensitive infectious disease includes vector-borne diseases such as malaria, whose transmission potential is expected to increase because of enhanced climatic suitability for the mosquito vector in Asia, sub-Saharan Africa, and South America. Climatic suitability for the mosquitoes that can carry dengue, Zika, and chikungunya is also likely to increase, facilitating further increases in the geographic range and longer transmission seasons, and raising concern for expansion of these diseases into temperate zones, particularly under higher greenhouse gas emission scenarios. Early spring temperatures in 2018 seem to have contributed to the early onset and extensive West Nile virus outbreak in Europe, a pathogen expected to expand further beyond its current distribution, due to a warming climate. As for tick-borne diseases, climate change is projected to continue to contribute to the spread of Lyme disease and tick-borne encephalitis, particularly in North America and Europe. Schistosomiasis is a water-borne disease and public health concern in Africa, Latin America, the Middle East, and Southeast Asia; climate change is anticipated to change its distribution, with both expansions and contractions expected. Other water-borne diseases that cause diarrheal diseases have declined significantly over the last decades owing to socioeconomic development and public health measures but changes in climate can reverse some of these positive developments. Weather and climate events, population movement, land use changes, urbanization, global trade, and other drivers can catalyze a succession of secondary events that can lead to a range of health impacts, including infectious disease outbreaks. These cascading risk pathways of causally connected events can result in large-scale outbreaks and affect society at large. We review climatic and other cascading drivers of infectious disease with projections under different climate change scenarios. Supplementary file1 (MP4 328467 KB).

Urban spatial risk prediction and optimization analysis of POI based on deep learning from the perspective of an epidemic

From an epidemiological perspective, previous research on COVID-19 has generally been based on classical statistical analyses. As a result, spatial information is often not used effectively. This paper uses image-based neural networks to explore the relationship between urban spatial risk and the distribution of infected populations, and the design of urban facilities. To achieve this objective, we use spatio-temporal data of people infected with new coronary pneumonia prior to 28 February 2020 in Wuhan. We then use kriging, which is a method of spatial interpolation, as well as core density estimation technology to establish the epidemic heat distribution on fine grid units. We further evaluate the influence of nine major spatial risk factors, including the distribution of agencies, hospitals, park squares, sports fields, banks and hotels, by testing them for significant positive correlation with the distribution of the epidemic. The weights of these spatial risk factors are used for training Generative Adversarial Network (GAN) models, which predict the distribution of cases in a given area. The input image for the machine learning model is a city plan converted by public infrastructures, and the output image is a map of urban spatial risk factors in the given area. The results of the trained model demonstrate that optimising the relevant point of interests (POI) in urban areas to effectively control potential risk factors can aid in managing the epidemic and preventing it from dispersing further.

Development and Use of a Reproducible Framework for Spatiotemporal Climatic Risk Assessment and its Association with Decadal Trend of Dengue in India

Introduction:

The study aimed to develop a reproducible, open-source, and scalable framework for extracting climate data from satellite imagery, understanding dengue's decadal trend in India, and estimating the relationship between dengue occurrence and climatic factors.

Materials and Methods:

A framework was developed in the Open Source Software, and it was empirically tested using reported annual dengue occurrence data in India during 2010–2019. Census 2011 and population projections were used to calculate incidence rates. Zonal statistics were performed to extract climate parameters. Correlation coefficients were calculated to estimate the relationship of dengue with the annual average of daily mean and minimum temperature and rainy days.

Results:

Total 818,973 dengue cases were reported from India, with median annual incidence of 6.57 per lakh population; it was high in 2019 and 2017 (11.80 and 11.55 per lakh) and the Southern region (8.18 per lakh). The highest median annual dengue incidence was observed in Punjab (24.49 per lakh). Daily climatic data were extracted from 1164 coordinate locations across the country for the decadal period (4,249,734 observations). The annual average of daily temperature and rainy days positively correlated with dengue in India (r = 0.31 and 0.06, at P < 0.01 and 0.30, respectively).

Conclusion:

The study provides a reproducible algorithm for bulk climatic data extraction from research-level satellite imagery. Infectious disease models can be used to understand disease epidemiology and strengthen disease surveillance in the country.

Data-driven methods for dengue prediction and surveillance using real-world and Big Data: A systematic review

Background

Traditionally, dengue surveillance is based on case reporting to a central health agency. However, the delay between a case and its notification can limit the system responsiveness. Machine learning methods have been developed to reduce the reporting delays and to predict outbreaks, based on non-traditional and non-clinical data sources. The aim of this systematic review was to identify studies that used real-world data, Big Data and/or machine learning methods to monitor and predict dengue-related outcomes.

Methodology/Principal findings

We performed a search in PubMed, Scopus, Web of Science and grey literature between January 1, 2000 and August 31, 2020. The review (ID: CRD42020172472) focused on data-driven studies. Reviews, randomized control trials and descriptive studies were not included. Among the 119 studies included, 67% were published between 2016 and 2020, and 39% used at least one novel data stream. The aim of the included studies was to predict a dengue-related outcome (55%), assess the validity of data sources for dengue surveillance (23%), or both (22%). Most studies (60%) used a machine learning approach. Studies on dengue prediction compared different prediction models, or identified significant predictors among several covariates in a model. The most significant predictors were rainfall (43%), temperature (41%), and humidity (25%). The two models with the highest performances were Neural Networks and Decision Trees (52%), followed by Support Vector Machine (17%). We cannot rule out a selection bias in our study because of our two main limitations: we did not include preprints and could not obtain the opinion of other international experts.

Conclusions/Significance

Combining real-world data and Big Data with machine learning methods is a promising approach to improve dengue prediction and monitoring. Future studies should focus on how to better integrate all available data sources and methods to improve the response and dengue management by stakeholders.

Dengue is one of the most important arbovirus infections in the world and its public health, societal and economic burden is increasing. Although the majority of dengue cases are asymptomatic or mild, severe disease forms can lead to death. For this reason, early diagnosis and monitoring of dengue are crucial to decrease mortality. However, most endemic regions still rely on traditional monitoring methods, despite the growing availability of novel data sources and data-driven methods based on real-world data, Big Data, and machine learning algorithms. In this systematic review, we identified and analyzed studies that used these novel approaches for dengue monitoring and/or prediction. We found that novel data streams, such as Internet search engines and social media platforms, and machine learning methods can be successfully used to improve dengue management, but are still vastly ignored in real life. These approaches should be combined with traditional methods to help stakeholders better prepare for each outbreak and improve early responsiveness.

Analysis and prediction of hand, foot and mouth disease incidence in China using Random Forest and XGBoost

Hand, foot and mouth disease (HFMD) is an increasingly serious public health problem, and it has caused an outbreak in China every year since 2008. Predicting the incidence of HFMD and analyzing its influential factors are of great significance to its prevention. Now, machine learning has shown advantages in infectious disease models, but there are few studies on HFMD incidence based on machine learning that cover all the provinces in mainland China. In this study, we proposed two different machine learning algorithms, Random Forest and eXtreme Gradient Boosting (XGBoost), to perform our analysis and prediction. We first used Random Forest to examine the association between HFMD incidence and potential influential factors for 31 provinces in mainland China. Next, we established Random Forest and XGBoost prediction models using meteorological and social factors as the predictors. Finally, we applied our prediction models in four different regions of mainland China and evaluated the performance of them. Our results show that: 1) Meteorological factors and social factors jointly affect the incidence of HFMD in mainland China. Average temperature and population density are the two most significant influential factors; 2) Population flux has different delayed effect in affecting HFMD incidence in different regions. From a national perspective, the model using population flux data delayed for one month has better prediction performance; 3) The prediction capability of XGBoost model was better than that of Random Forest model from the overall perspective. XGBoost model is more suitable for predicting the incidence of HFMD in mainland China.

Dengue disease dynamics are modulated by the combined influences of precipitation and landscape: A machine learning approach

BACKGROUND

Dengue is an endemic vector-borne disease influenced by environmental factors such as landscape and climate. Previous studies separately assessed the effects of landscape and climate factors on mosquito occurrence and dengue incidence. However, both factors concurrently coexist in time and space and can interact, affecting mosquito development and dengue disease transmission. For example, eggs laid in a suitable environment can hatch after being submerged in rain water. It has been difficult for conventional statistical modeling approaches to demonstrate these combined influences due to mathematical constraints.

OBJECTIVES

To investigate the combined influences of landscape and climate factors on mosquito occurrence and dengue incidence.

METHODS

Entomological, epidemiological, and landscape data from the rainy season (July-December) were obtained from respective government agencies in Metropolitan Manila, Philippines, from 2012 to 2014. Temperature, precipitation and vegetation data were obtained through remote sensing. A random forest algorithm was used to select the landscape and climate variables. Afterward, using the identified key variables, a model-based (MOB) recursive partitioning was implemented to test the combined influences of landscape and climate factors on ovitrap index (vector mosquito occurrence) and dengue incidence.

RESULTS

The MOB recursive partitioning for ovitrap index indicated a high sensitivity of vector mosquito occurrence on environmental conditions generated by a combination of high residential density areas with low precipitation. Moreover, the MOB recursive partitioning indicated high sensitivity of dengue incidence to the effects of precipitation in areas with high proportions of residential density and commercial areas.

CONCLUSIONS

Dengue dynamics are not solely influenced by individual effects of either climate or landscape, but rather by their synergistic or combined effects. The presented findings have the potential to target vector surveillance in areas identified as suitable for mosquito occurrence under specific climatic conditions and may be relevant as part of urban planning strategies to control dengue.

Dengue models based on machine learning techniques: A systematic literature review

BACKGROUND

Dengue modeling is a research topic that has increased in recent years. Early prediction and decision-making are key factors to control dengue. This Systematic Literature Review (SLR) analyzes three modeling approaches of dengue: diagnostic, epidemic, intervention. These approaches require models of prediction, prescription and optimization. This SLR establishes the state-of-the-art in dengue modeling, using machine learning, in the last years.

METHODS

Several databases were selected to search the articles. The selection was made based on Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology. Sixty-four articles were obtained and analyzed to describe their strengths and limitations. Finally, challenges and opportunities for research on machine-learning for dengue modeling were identified.

RESULTS

Logistic regression was the most used modeling approach for the diagnosis of dengue (59.1%). The analysis of the epidemic approach showed that linear regression (17.4%) is the most used technique within the spatial analysis. Finally, the most used intervention modeling is General Linear Model with 70%.

CONCLUSIONS

We conclude that cause-effect models may improve diagnosis and understanding of dengue. Models that manage uncertainty can also be helpful, because of low data-quality in healthcare. Finally, decentralization of data, using federated learning, may decrease computational costs and allow model building without compromising data security.

The Complex Epidemiological Relationship between Flooding Events and Human Outbreaks of Mosquito-Borne Diseases: A Scoping Review

BACKGROUND

Climate change is expected to increase the frequency of flooding events. Although rainfall is highly correlated with mosquito-borne diseases (MBD) in humans, less research focuses on understanding the impact of flooding events on disease incidence. This lack of research presents a significant gap in climate change-driven disease forecasting.

OBJECTIVES

We conducted a scoping review to assess the strength of evidence regarding the potential relationship between flooding and MBD and to determine knowledge gaps.

METHODS

PubMed, Embase, and Web of Science were searched through 31 December 2020 and supplemented with review of citations in relevant publications. Studies on rainfall were included only if the operationalization allowed for distinction of unusually heavy rainfall events. Data were abstracted by disease (dengue, malaria, or other) and stratified by post-event timing of disease assessment. Studies that conducted statistical testing were summarized in detail.

RESULTS

From 3,008 initial results, we included 131 relevant studies (dengue n = 45 , malaria n = 61 , other MBD n = 49 ). Dengue studies indicated short-term (< 1  month ) decreases and subsequent (1-4 month) increases in incidence. Malaria studies indicated post-event incidence increases, but the results were mixed, and the temporal pattern was less clear. Statistical evidence was limited for other MBD, though findings suggest that human outbreaks of Murray Valley encephalitis, Ross River virus, Barmah Forest virus, Rift Valley fever, and Japanese encephalitis may follow flooding.

DISCUSSION

Flooding is generally associated with increased incidence of MBD, potentially following a brief decrease in incidence for some diseases. Methodological inconsistencies significantly limit direct comparison and generalizability of study results. Regions with established MBD and weather surveillance should be leveraged to conduct multisite research to a) standardize the quantification of relevant flooding, b) study nonlinear relationships between rainfall and disease, c) report outcomes at multiple lag periods, and d) investigate interacting factors that modify the likelihood and severity of outbreaks across different settings. https://doi.org/10.1289/EHP8887.

Increased coverage and high confidence in suspect screening of emerging contaminants in global environmental samples

Suspect screening using liquid chromatography with high resolution mass spectrometry provides an opportunity for expanding the detection coverage of emerging contaminants in the environment. Screening workflows may suffer from high frequency of false positives or insufficient confidence in the identification of compounds; however, stringent criteria could lead to high false negatives. A workflow must have a balanced criteria, both selective and sensitive, to be able to identify real features without missing low abundant features traceable to analytes of interest. A highly selective (87%) and sensitive (97%) workflow was developed by characterizing the occurrence of contaminants in wastewater and surface water from Hong Kong, India, Philippines, Sweden, Switzerland, and the U.S. Sixty-eight contaminants were identified and confirmed with reference standards, including pharmaceuticals, pesticides, and industrial chemicals. The antimicrobials metronidazole, clindamycin, linezolid, and rifaximin were detected. Notably, antifungal compounds were detected in samples from six countries, with levels up to 1380 ng/L. Amoxicillin transformation products, penilloic acid (285-8047 ng/L) and penicilloic acid (107 ng/L), were confirmed for the first time with reference standards in wastewater samples from India, Sweden, and U.S. This workflow provides an efficient approach to broad-scale identification of emerging contaminants using publicly-available databases for suspect screening and prioritization.

Emergency room visits for childhood atopic dermatitis are associated with floods?

Floods are known to increase the level of allergens such as molds in the environments. Under climate change, the frequency of floods could be increased, which highlights the importance of understanding the impacts of floods on atopic diseases. However, there was a lack of studies. This study examines whether floods induce attacks of childhood atopic dermatitis (AD). A retrospective population-based study was conducted in Taiwan Island using Taiwan's National Health Insurance Research Database. Emergency room (ER) visits for AD were identified among children aged 0-12 years. Weekly data of flood occurrence, number of flood sites, temperature, and air pollution were obtained for each township of the identified cases. A time-stratified case-crossover design was used. The relationship between ER visits for AD and floods was assessed by conditional logistic regression, adjusting for weekly mean temperature, PM_2.5 and NO₂. There were a total of 55,488 ER visits due to AD during the study period. Such visits increased when flood occurred, and then declined. The effects of floods at the week of flood remained robust, with OR of 1.14 (95% CI = 1.01-1.28) for flood occurrence and 1.31 (95% CI = 1.10-1.55) for the number of flood sites, after adjusting for covariates. Such effects were slightly higher in boys and children aged 1-12 years. This study demonstrated the impact of floods on flare-up of childhood AD, and the effect was most prominently at the week of flood. Healthcare workers should be alarmed for potential increase of AD flare ups after flood events.

Software Defect Prediction for Healthcare Big Data: An Empirical Evaluation of Machine Learning Techniques

Software defect prediction (SDP) in the initial period of the software development life cycle (SDLC) remains a critical and important assignment. SDP is essentially studied during few last decades as it leads to assure the quality of software systems. The quick forecast of defective or imperfect artifacts in software development may serve the development team to use the existing assets competently and more effectively to provide extraordinary software products in the given or narrow time. Previously, several canvassers have industrialized models for defect prediction utilizing machine learning (ML) and statistical techniques. ML methods are considered as an operative and operational approach to pinpoint the defective modules, in which moving parts through mining concealed patterns amid software metrics (attributes). ML techniques are also utilized by several researchers on healthcare datasets. This study utilizes different ML techniques software defect prediction using seven broadly used datasets. The ML techniques include the multilayer perceptron (MLP), support vector machine (SVM), decision tree (J48), radial basis function (RBF), random forest (RF), hidden Markov model (HMM), credal decision tree (CDT), K-nearest neighbor (KNN), average one dependency estimator (A1DE), and Naïve Bayes (NB). The performance of each technique is evaluated using different measures, for instance, relative absolute error (RAE), mean absolute error (MAE), root mean squared error (RMSE), root relative squared error (RRSE), recall, and accuracy. The inclusive outcome shows the best performance of RF with 88.32% average accuracy and 2.96 rank value, second-best performance is achieved by SVM with 87.99% average accuracy and 3.83 rank values. Moreover, CDT also shows 87.88% average accuracy and 3.62 rank values, placed on the third position. The comprehensive outcomes of research can be utilized as a reference point for new research in the SDP domain, and therefore, any assertion concerning the enhancement in prediction over any new technique or model can be benchmarked and proved.

Prediction of dengue outbreak in Selangor Malaysia using machine learning techniques

Dengue fever is a mosquito-borne disease that affects nearly 3.9 billion people globally. Dengue remains endemic in Malaysia since its outbreak in the 1980's, with its highest concentration of cases in the state of Selangor. Predictors of dengue fever outbreaks could provide timely information for health officials to implement preventative actions. In this study, five districts in Selangor, Malaysia, that demonstrated the highest incidence of dengue fever from 2013 to 2017 were evaluated for the best machine learning model to predict Dengue outbreaks. Climate variables such as temperature, wind speed, humidity and rainfall were used in each model. Based on results, the SVM (linear kernel) exhibited the best prediction performance (Accuracy = 70%, Sensitivity = 14%, Specificity = 95%, Precision = 56%). However, the sensitivity for SVM (linear) for the testing sample increased up to 63.54% compared to 14.4% for imbalanced data (original data). The week-of-the-year was the most important predictor in the SVM model. This study exemplifies that machine learning has respectable potential for the prediction of dengue outbreaks. Future research should consider boosting, or using, nature inspired algorithms to develop a dengue prediction model.

Prediction for Global Peste des Petits Ruminants Outbreaks Based on a Combination of Random Forest Algorithms and Meteorological Data

Peste des Petits Ruminants (PPR) is an acute and highly contagious transboundary disease caused by the PPR virus (PPRV). The virus infects goats, sheep and some wild relatives of small domestic ruminants, such as antelopes. PPR is listed by the World Organization for Animal Health as an animal disease that must be reported promptly. In this paper, PPR outbreak data combined with WorldClim database meteorological data were used to build a PPR prediction model. Using feature selection methods, eight sets of features were selected: bio3, bio10, bio15, bio18, prec7, prec8, prec12, and alt for modeling. Then different machine learning algorithms were used to build models, among which the random forest (RF) algorithm was found to have the best modeling effect. The ACC value of prediction accuracy for the model on the training set can reach 99.10%, while the ACC on the test sets was 99.10%. Therefore, RF algorithms and eight features were finally selected to build the model in order to build the online prediction system. In addition, we adopt single-factor modeling and correlation analysis of modeling variables to explore the impact of each variable on modeling results. It was found that bio18 (the warmest quarterly precipitation), prec7 (the precipitation in July), and prec8 (the precipitation in August) contributed significantly to the model, and the outbreak of the epidemic may have an important relationship with precipitation. Eventually, we used the final qualitative prediction model to establish a global online prediction system for the PPR epidemic.

Improved Transferability of Data-Driven Damage Models Through Sample Selection Bias Correction

Damage models for natural hazards are used for decision making on reducing and transferring risk. The damage estimates from these models depend on many variables and their complex sometimes nonlinear relationships with the damage. In recent years, data-driven modeling techniques have been used to capture those relationships. The available data to build such models are often limited. Therefore, in practice it is usually necessary to transfer models to a different context. In this article, we show that this implies the samples used to build the model are often not fully representative for the situation where they need to be applied on, which leads to a "sample selection bias." In this article, we enhance data-driven damage models by applying methods, not previously applied to damage modeling, to correct for this bias before the machine learning (ML) models are trained. We demonstrate this with case studies on flooding in Europe, and typhoon wind damage in the Philippines. Two sample selection bias correction methods from the ML literature are applied and one of these methods is also adjusted to our problem. These three methods are combined with stochastic generation of synthetic damage data. We demonstrate that for both case studies, the sample selection bias correction techniques reduce model errors, especially for the mean bias error this reduction can be larger than 30%. The novel combination with stochastic data generation seems to enhance these techniques. This shows that sample selection bias correction methods are beneficial for damage model transfer.

Cholera Risk: A Machine Learning Approach Applied to Essential Climate Variables

Oceanic and coastal ecosystems have undergone complex environmental changes in recent years, amid a context of climate change. These changes are also reflected in the dynamics of water-borne diseases as some of the causative agents of these illnesses are ubiquitous in the aquatic environment and their survival rates are impacted by changes in climatic conditions. Previous studies have established strong relationships between essential climate variables and the coastal distribution and seasonal dynamics of the bacteria Vibrio cholerae, pathogenic types of which are responsible for human cholera disease. In this study we provide a novel exploration of the potential of a machine learning approach to forecast environmental cholera risk in coastal India, home to more than 200 million inhabitants, utilising atmospheric, terrestrial and oceanic satellite-derived essential climate variables. A Random Forest classifier model is developed, trained and tested on a cholera outbreak dataset over the period 2010–2018 for districts along coastal India. The random forest classifier model has an Accuracy of 0.99, an F1 Score of 0.942 and a Sensitivity score of 0.895, meaning that 89.5% of outbreaks are correctly identified. Spatio-temporal patterns emerged in terms of the model’s performance based on seasons and coastal locations. Further analysis of the specific contribution of each Essential Climate Variable to the model outputs shows that chlorophyll-a concentration, sea surface salinity and land surface temperature are the strongest predictors of the cholera outbreaks in the dataset used. The study reveals promising potential of the use of random forest classifiers and remotely-sensed essential climate variables for the development of environmental cholera-risk applications. Further exploration of the present random forest model and associated essential climate variables is encouraged on cholera surveillance datasets in other coastal areas affected by the disease to determine the model’s transferability potential and applicative value for cholera forecasting systems.

Weekly dengue forecasts in Iquitos, Peru; San Juan, Puerto Rico; and Singapore

BACKGROUND

Predictive models can serve as early warning systems and can be used to forecast future risk of various infectious diseases. Conventionally, regression and time series models are used to forecast dengue incidence, using dengue surveillance (e.g., case counts) and weather data. However, these models may be limited in terms of model assumptions and the number of predictors that can be included. Machine learning (ML) methods are designed to work with a large number of predictors and thus offer an appealing alternative. Here, we compared the performance of ML algorithms with that of regression models in predicting dengue cases and outbreaks from 4 to up to 12 weeks in advance. Many countries lack sufficient health surveillance infrastructure, as such we evaluated the contribution of dengue surveillance and weather data on the predictive power of these models.

METHODS

We developed ML, regression, and time series models to forecast weekly dengue case counts and outbreaks in Iquitos, Peru; San Juan, Puerto Rico; and Singapore from 1990-2016. Forecasts were generated using available weekly dengue surveillance, and weather data. We evaluated the agreement between model forecasts and actual dengue observations using Mean Absolute Error and Matthew's Correlation Coefficient (MCC).

RESULTS

For near term predictions of weekly case counts and when using surveillance data, ML models had 21% and 33% less error than regression and time series models respectively. However, using weather data only, ML models did not demonstrate a practical advantage. When forecasting weekly dengue outbreaks 12 weeks in advance, ML models achieved a maximum MCC of 0.61.

CONCLUSIONS

Our results identified 2 scenarios when ML models are advantageous over regression model: 1) predicting dengue weekly case counts 4 weeks ahead when dengue surveillance data are available and 2) predicting weekly dengue outbreaks 12 weeks ahead when dengue surveillance data are unavailable. Given the advantages of ML models, dengue early warning systems may be improved by the inclusion of these models.

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

The robust estimate and forecast capability of random forests (RF) has been widely recognized, however this ensemble machine learning method has not been widely used in mosquito-borne disease forecasting. In this study, two sets of RF models were developed at the national (pooled department-level data) and department level in Colombia to predict weekly dengue cases for 12-weeks ahead. A pooled national model based on artificial neural networks (ANN) was also developed and used as a comparator to the RF models. The various predictors included historic dengue cases, satellite-derived estimates for vegetation, precipitation, and air temperature, as well as population counts, income inequality, and education. Our RF model trained on the pooled national data was more accurate for department-specific weekly dengue cases estimation compared to a local model trained only on the department's data. Additionally, the forecast errors of the national RF model were smaller to those of the national pooled ANN model and were increased with the forecast horizon increasing from one-week-ahead (mean absolute error, MAE: 9.32) to 12-weeks ahead (MAE: 24.56). There was considerable variation in the relative importance of predictors dependent on forecast horizon. The environmental and meteorological predictors were relatively important for short-term dengue forecast horizons while socio-demographic predictors were relevant for longer-term forecast horizons. This study demonstrates the potential of RF in dengue forecasting with a feasible approach of using a national pooled model to forecast at finer spatial scales. Furthermore, including sociodemographic predictors is likely to be helpful in capturing longer-term dengue trends.

Assessment of the outbreak risk, mapping and infection behavior of COVID-19: Application of the autoregressive integrated-moving average (ARIMA) and polynomial models

Infectious disease outbreaks pose a significant threat to human health worldwide. The outbreak of pandemic coronavirus disease 2019 (COVID-19) has caused a global health emergency. Thus, identification of regions with high risk for COVID-19 outbreak and analyzing the behaviour of the infection is a major priority of the governmental organizations and epidemiologists worldwide. The aims of the present study were to analyze the risk factors of coronavirus outbreak for identifying the areas having high risk of infection and to evaluate the behaviour of infection in Fars Province, Iran. A geographic information system (GIS)-based machine learning algorithm (MLA), support vector machine (SVM), was used for the assessment of the outbreak risk of COVID-19 in Fars Province, Iran whereas the daily observations of infected cases were tested in the-polynomial and the autoregressive integrated moving average (ARIMA) models to examine the patterns of virus infestation in the province and in Iran. The results of the disease outbreak in Iran were compared with the data for Iran and the world. Sixteen effective factors were selected for spatial modelling of outbreak risk. The validation outcome reveals that SVM achieved an AUC value of 0.786 (March 20), 0.799 (March 29), and 86.6 (April 10) that displays a good prediction of outbreak risk change detection. The results of the third-degree polynomial and ARIMA models in the province revealed an increasing trend with an evidence of turning, demonstrating extensive quarantines has been effective. The general trends of virus infestation in Iran and Fars Province were similar, although a more volatile growth of the infected cases is expected in the province. The results of this study might assist better programming COVID-19 disease prevention and control and gaining sorts of predictive capability would have wide-ranging benefits.

Assessment of vector-host-pathogen relationships using data mining and machine learning

Infectious diseases, including vector-borne diseases transmitted by arthropods, are a leading cause of morbidity and mortality worldwide. In the era of big data, addressing broad-scale, fundamental questions regarding the complex dynamics of these diseases will increasingly require the integration of diverse datasets to produce new biological knowledge. This review provides a current snapshot of the systematic assessment of the relationships between microbial pathogens, arthropod vectors and mammalian hosts using data mining and machine learning. We employ PRISMA to identify 32 key papers relevant to this topic. Our analysis shows an increasing use of data mining and machine learning tasks and techniques, including prediction, classification, clustering, association rules mining, and deep learning, over the last decade. However, it also reveals a number of critical challenges in applying these to the study of vector-host-pathogen interactions at various systems biology levels. Here, relevant studies, current limitations and future directions are discussed. Furthermore, the quality of data in relevant papers was assessed using the FAIR (Findable, Accessible, Interoperable, Reusable) compliance criteria to evaluate and encourage reproducibility and shareability of research outcomes. Although shortcomings in their application remain, data mining and machine learning have significant potential to break new ground in understanding fundamental aspects of vector-host-pathogen relationships and their application in this field should be encouraged. In particular, while predictive modeling, feature engineering and supervised machine learning are already being used in the field, other data mining and machine learning methods such as deep learning and association rules analysis lag behind and should be implemented in combination with established methods to accelerate hypothesis and knowledge generation in the domain.

Fine-scale population genetic structure of dengue mosquito vector, Aedes aegypti, in Metropolitan Manila, Philippines

Dengue is a highly endemic disease in Southeast Asia and is transmitted primarily by the mosquito, Aedes aegypti. The National Capital Region (NCR) of the Philippines, or Metropolitan Manila, is a highly urbanized area that is greatly affected by this arboviral disease. Urbanization has been shown to increase the dispersal of this mosquito vector. For this reason, we conducted a fine-scale population genetic study of Ae. aegypti in this region. We collected adult Ae. aegypti mosquitoes (n = 526 individuals) within the region (n = 21 study areas) and characterized the present population structure and the genetic relatedness among mosquito populations. We genotyped 11 microsatellite loci from all sampled mosquito individuals and analyzed their genetic diversity, differentiation and structure. The results revealed low genetic differentiation across mosquito populations which suggest high gene flow and/or weak genetic drift among mosquito populations. Bayesian analysis indicated multiple genetic structures (K = 3-6), with no clear genetically distinct population structures. This result implies the passive or long-distance dispersal capability nature Ae. aegypti possibly through human-mediated transportation. The constructed dendrogram in this study describes the potential passive dispersal patterns across Metropolitan Manila. Furthermore, spatial autocorrelation analysis showed the limited and active dispersal capability (<1km) of the mosquito vector. Our findings are consistent with previous studies that investigated the genetic structure and dual (active and passive) dispersal capability of Ae. aegypti in a fine-scale highly urbanized area.

Statistical methods for predicting tuberculosis incidence based on data from Guangxi, China

Background

Tuberculosis (TB) remains a serious public health problem with substantial financial burden in China. The incidence of TB in Guangxi province is much higher than that in the national level, however, there is no predictive study of TB in recent years in Guangxi, therefore, it is urgent to construct a model to predict the incidence of TB, which could provide help for the prevention and control of TB.

Methods

Box-Jenkins model methods have been successfully applied to predict the incidence of infectious disease. In this study, based on the analysis of TB incidence in Guangxi from January 2012 to June 2019, we constructed TB prediction model by Box-Jenkins methods, and used root mean square error (RMSE), mean absolute error (MAE) and mean absolute percentage error (MAPE) to test the performance and prediction accuracy of model.

Results

From January 2012 to June 2019, a total of 587,344 cases of TB were reported and 879 cases died in Guangxi. Based on TB incidence from January 2012 to December 2018, the SARIMA((2),0,(2))(0,1,0)₁₂ model was established, the AIC and SC of this model were 2.87 and 2.98, the fitting accuracy indexes, such as RMSE, MAE and MAPE were 0.98, 0.77 and 5.8 respectively; the prediction accuracy indexes, such as RMSE, MAE and MAPE were 0.62, 0.45 and 3.77, respectively. Based on the SARIMA((2),0,(2))(0,1,0)₁₂ model, we predicted the TB incidence in Guangxi from July 2019 to December 2020.

Conclusions

This study filled the gap in the prediction of TB incidence in Guangxi in recent years. The established SARIMA((2),0,(2))(0,1,0)₁₂ model has high prediction accuracy and good prediction performance. The results suggested the change trend of TB incidence predicted by SARIMA((2),0,(2))(0,1,0)₁₂ model from July 2019 to December 2020 was similar to that in the previous two years, and TB incidence will experience slight decrease, the predicted results can provide scientific reference for the prevention and control of TB in Guangxi, China.

High relative humidity might trigger the occurrence of the second seasonal peak of dengue in the Philippines

BACKGROUND

Dengue in some regions has a bimodal seasonal pattern, with a first big seasonal peak followed by a second small seasonal peak. The factors associated with the second small seasonal peak remain unclear.

METHODS

Monthly data on dengue cases in the Philippines and its 17 regions from 2008 to 2017 were collected and underwent a time series seasonal decomposition analysis. The associations of monthly average mean temperature, average relative humidity, and total rainfall with dengue in 19 provinces were assessed with a generalized additive model. Logistic regression and a classification and regression tree (CART) model were used to identify the factors associated with the second seasonal peak of dengue.

RESULTS

Dengue incidence rate in the Philippines increased substantially in the period 2013-2017, particularly for the regions in south Philippines. Dengue peaks in south Philippines predominantly occurred in August, with the peak in the national capital region (NCR) (i.e., Metropolitan Manila) occurring in September. The association between mean temperature and dengue appeared J-shaped or upside-down-V-shaped, and the association between relative humidity (or rainfall) and dengue was heterogeneous across different provinces (e.g., J shape, reverse J shape, or upside-down V shape, etc). Relative humidity was the only factor associated with the second seasonal peak of dengue (odds ratio: 1.144; 95% confidence interval: 1.023-1.279; threshold: 77%).

CONCLUSIONS

Dengue control and prevention resources are increasingly required in regions beyond the NCR, and relative humidity can be used as a predictor of the second seasonal peak of dengue in the Philippines.

Influenza activity prediction using meteorological factors in a warm temperate to subtropical transitional zone, Eastern China

Influenza activity is subject to environmental factors. Accurate forecasting of influenza epidemics would permit timely and effective implementation of public health interventions, but it remains challenging. In this study, we aimed to develop random forest (RF) regression models including meterological factors to predict seasonal influenza activity in Jiangsu provine, China. Coefficient of determination (R2) and mean absolute percentage error (MAPE) were employed to evaluate the models' performance. Three RF models with optimum parameters were constructed to predict influenza like illness (ILI) activity, influenza A and B (Flu-A and Flu-B) positive rates in Jiangsu. The models for Flu-B and ILI presented excellent performance with MAPEs <10%. The predicted values of the Flu-A model also matched the real trend very well, although its MAPE reached to 19.49% in the test set. The lagged dependent variables were vital predictors in each model. Seasonality was more pronounced in the models for ILI and Flu-A. The modification effects of the meteorological factors and their lagged terms on the prediction accuracy differed across the three models, while temperature always played an important role. Notably, atmospheric pressure made a major contribution to ILI and Flu-B forecasting. In brief, RF models performed well in influenza activity prediction. Impacts of meteorological factors on the predictive models for influenza activity are type-specific.

Prediction for global African swine fever outbreaks based on a combination of random forest algorithms and meteorological data

African swine fever (ASF) is a virulent infectious disease of pigs. As there is no effective vaccine and treatment method at present, it poses a great threat to the pig industry once it breaks out. In this paper, we used ASF outbreak data and the WorldClim database meteorological data and selected the CfsSubset Evaluator-Best First feature selection method combined with the random forest algorithms to construct an African swine fever outbreak prediction model. Subsequently, we also established a test set for data other than modelling, and the accuracy accuracy value range of the model on the independent test set was 76.02%-84.64%, which indicated that the modelling effect was better and the prediction accuracy was higher than previous estimates. In addition, logistic regression analysis was conducted on 12 features used for modelling and the ROC curves were drawn. The results showed that the bio14 features (precipitation of driest month) had the largest contribution to the outbreak of ASF, and it was speculated that the outbreak of the epidemic was significantly related to precipitation. Finally, we used this qualitative prediction model to build a global online prediction system for ASF outbreaks, in the hope that this study will help to decision-makers who can then take the relevant prevention and control measures in order to prevent the further spread of future epidemics of the disease.

Predictive Models for the Medical Diagnosis of Dengue: A Case Study in Paraguay

Early diagnosis of dengue continues to be a concern for public health in countries with a high incidence of this disease. In this work, we compared two machine learning techniques: artificial neural networks (ANN) and support vector machines (SVM) as assistance tools for medical diagnosis. The performance of classification models was evaluated in a real dataset of patients with a previous diagnosis of dengue extracted from the public health system of Paraguay during the period 2012-2016. The ANN multilayer perceptron achieved better results with an average of 96% accuracy, 96% sensitivity, and 97% specificity, with low variation in thirty different partitions of the dataset. In comparison, SVM polynomial obtained results above 90% for accuracy, sensitivity, and specificity.

Detection of Wolbachia in field-collected Aedes aegypti mosquitoes in metropolitan Manila, Philippines

BACKGROUND

Recent reports reveal the presence of Wolbachia in Ae. aegypti. Our study presents additional support for Wolbachia infection in Ae. aegypti by screening field-collected adult mosquitoes using two Wolbachia-specific molecular makers.

METHODS

A total of 672 Ae. aegypti adult mosquitoes were collected from May 2014 to January 2015 in Metropolitan Manila. Each individual sample was processed and screened for the presence of Wolbachia by selected markers, Wolbachia-specific 16S rDNA and its surface protein (wsp), under optimized PCR conditions and sequenced.

RESULTS

Totals of 113 (16.8%) and 89 (13.2%) individual mosquito samples were determined to be infected with Wolbachia using the wsp and 16S rDNA markers, respectively. The Ae. aegpyti wsp sample sequences were similar or identical to five known Wolbachia strains belonging to supergroups A and B while the majority of 16S rDNA sample sequences were similar to strains belonging to supergroup B. Overall, 80 (11.90%) individual mosquito samples showed positive amplifications in both markers and 69% showed congruence in supergroup identification (supergroup B).

CONCLUSIONS

By utilizing two Wolbachia-specific molecular makers, our study demonstrated the presence of Wolbachia from individual Ae. aegypti samples. Our results showed a low Wolbachia infection rate and inferred the detected strains belong to either supergroups A and B.

Spatiotemporal patterns and climatic drivers of severe dengue in Thailand

OBJECTIVES

The burden of dengue fever in Thailand is considerable, yet there are few large-scale studies exploring the drivers of transmission. This study aimed to investigate the spatiotemporal patterns and climatic drivers of severe dengue in Thailand.

METHODS

Geographic Information System (GIS) techniques and spatial cluster analysis were used to visualize the spatial distribution and detect high-risk clusters of severe dengue in 76 provinces of Thailand from January 1999 to December 2014. The seasonal patterns of severe dengue cases in different provinces were identified. A two-stage modelling approach combining a generalized linear model with a distributed lag non-linear model was used to quantify the effects of monthly mean temperature and relative humidity on the occurrence of severe dengue cases in 51 provinces of Thailand.

RESULTS

Significant severe dengue clustering was detected, especially during epidemic years, and the location of these clusters showed substantial inter-annual variation. Severe dengue cases in Northern and Northeastern Thailand peaked in June to August and this pattern was stable across the study period, whereas the seasonality of severe dengue cases in other regions (especially Central Thailand) was less predictable. The risk of the occurrence of severe dengue cases increased with an increase in mean temperature in Northeastern Thailand, Central Thailand, and Southern Thailand, with peaks occurring between 24 °C to 30 °C in Northeastern Thailand and 27 °C to 29 °C in Southern Thailand West Coast, respectively. Relative humidity significantly affected the occurrence of severe dengue cases in Northeastern and Central Thailand, with optimal ranges observed for each region.

CONCLUSIONS

Our findings substantiate the potential for developing climate-based dengue early warning systems for Thailand, and have implications for informing pre-emptive vector control.

Spatial epidemiological approaches to inform leptospirosis surveillance and control: A systematic review and critical appraisal of methods

Leptospirosis is a global zoonotic disease that the transmission is driven by complex geographical and temporal variation in demographics, animal hosts and socioecological factors. This results in complex challenges for the identification of high-risk areas. Spatial and temporal epidemiological tools could be used to support leptospirosis control programs, but the adequacy of its application has not been evaluated. We searched literature in six databases including PubMed, Web of Science, EMBASE, Scopus, SciELO and Zoological Record to systematically review and critically assess the use of spatial and temporal analytical tools for leptospirosis and to provide general framework for its application in future studies. We reviewed 115 articles published between 1930 and October 2018 from 41 different countries. Of these, 65 (56.52%) articles were on human leptospirosis, 39 (33.91%) on animal leptospirosis and 11 (9.5%) used data from both human and animal leptospirosis. Spatial analytical (n = 106) tools were used to describe the distribution of incidence/prevalence at various geographical scales (96.5%) and to explored spatial patterns to detect clustering and hot spots (33%). A total of 51 studies modelled the relationships of various variables on the risk of human (n = 31), animal (n = 17) and both human and animal infection (n = 3). Among those modelling studies, few studies had generated spatially structured models and predictive maps of human (n = 2/31) and animal leptospirosis (n = 1/17). In addition, nine studies applied time-series analytical tools to predict leptospirosis incidence. Spatial and temporal analytical tools have been greatly utilized to improve our understanding on leptospirosis epidemiology. Yet the quality of the epidemiological data, the selection of covariates and spatial analytical techniques should be carefully considered in future studies to improve usefulness of evidence as tools to support leptospirosis control. A general framework for the application of spatial analytical tools for leptospirosis was proposed.

Using Google Trends to Examine the Spatio-Temporal Incidence and Behavioral Patterns of Dengue Disease: A Case Study in Metropolitan Manila, Philippines

Dengue is a major public health concern and an economic burden in the Philippines. Despite the country's improved dengue surveillance, it still suffers from various setbacks and needs to be complemented with alternative approaches. Previous studies have demonstrated the potential of Internet-based surveillance such as Google Dengue Trends (GDT) in supplementing current epidemiological methods for predicting future dengue outbreaks and patterns. With this, our study has two objectives: (1) assess the temporal relationship of weekly GDT and dengue incidence in Metropolitan Manila from 2009⁻2014; and (2) examine the health-seeking behavior based on dengue-related search queries of the population. The study collated the population statistics and reported dengue cases in Metropolitan Manila from respective government agencies to calculate the dengue incidence (DI) on a weekly basis for the entire region and annually per city. Data processing of GDT and dengue incidence was performed by conducting an 'adjustment' and scaling procedures, respectively, and further analyzed for correlation and cross-correlation analyses using Pearson's correlation. The relative search volume of the term 'dengue' and top dengue-related search queries in Metropolitan Manila were obtained and organized from the Google Trends platform. Afterwards, a thematic analysis was employed, and word clouds were generated to examine the health behavior of the population. Results showed that weekly temporal GDT pattern are closely similar to the weekly DI pattern in Metropolitan Manila. Further analysis showed that GDT has a moderate and positive association with DI when adjusted or scaled, respectively. Cross-correlation analysis revealed a delayed effect where GDT leads DI by 1⁻2 weeks. Thematic analysis of dengue-related search queries indicated 5 categories namely; (a) dengue, (b) sign and symptoms of dengue, (c) treatment and prevention, (d) mosquito, and (e) other diseases. The majority of the search queries were classified in 'signs and symptoms' which indicate the health-seeking behavior of the population towards the disease. Therefore, GDT can be utilized to complement traditional disease surveillance methods combined with other factors that could potentially identify dengue hotspots and help in public health decisions.