Analysis of significant factors for dengue fever incidence prediction

Understanding local determinants of dengue: a geographically weighted panel regression approach in Yogyakarta, Indonesia

BACKGROUND

Dengue remains a major public health concern in tropical regions, including Yogyakarta, Indonesia. Understanding its spatiotemporal patterns and determinants is crucial for effective prevention strategies. This study explores the spatiotemporal determinants of dengue incidence and evaluates the spatial variability of predictors using a geographically weighted panel regression (GWPR) approach.

METHODS

This ecological study applied a spatiotemporal approach, analyzing dengue incidence across 78 sub-districts in Yogyakarta from 2017 to 2022. The dataset included meteorological variables (rainfall, temperature, humidity, wind speed, and atmospheric pressure), sociodemographic data (population density), and land-use characteristics (built-up areas, crops, trees, water bodies, and flooded vegetation). A GWPR model with a Fixed Exponential kernel was used to assess local variations in predictor influence.

RESULTS

The Fixed Exponential Kernel GWPR model showed strong explanatory power (Adjusted R2 = 0.516, RSS = 43,097.96, AIC = 28,447.38). Local R-Square values ranged from 0.25 (low-performing sub-districts) to 0.75 (high-performing sub-districts), indicating significant spatial heterogeneity. Sub-districts such as Pakem, Cangkringan, and Girimulyo exhibited high local R2 values (>0.75), indicating robust model performance, whereas Kalibawang showed lower values (<0.25), suggesting weaker predictive power. High-dengue-burden sub-districts, including Kasihan (0.743), Banguntapan (0.731), Sewon (0.716), Wonosari (0.623), and Wates (0.540), demonstrated stronger associations between dengue incidence and key predictors. In Wonosari, the most influential predictors were Rainfall Lag 1, Rainfall Lag 3, temperature, humidity, wind speed, atmospheric pressure, and land-use variables, while in Wates, significant predictors included Rainfall Lag 1, Rainfall Lag 3, atmospheric pressure, and land-use factors. Lower model performance in Sedayu and Kalibawang suggests the necessity of incorporating additional predictors such as sanitation conditions and vector control activities.

CONCLUSIONS

The GWPR model provides valuable insights into the spatiotemporal dynamics of dengue incidence, emphasizing the role of localized predictors. Spatially adaptive prevention strategies focusing on high-risk areas are essential for effective dengue control in Yogyakarta and similar tropical regions.

Predicting spatio-temporal dynamics of dengue using INLA (integrated nested laplace approximation) in Yogyakarta, Indonesia

INTRODUCTION

Dengue is a mosquito-borne disease caused by the dengue virus, primarily transmitted by Aedes aegypti and Aedes albopictus. Its incidence fluctuates due to spatial and temporal factors, necessitating robust modeling approaches for prediction and risk mapping.

OBJECTIVES

This study aims to develop a spatio-temporal Bayesian model for predicting dengue incidence, integrating climatic, sociodemographic, and environmental factors to improve outbreak forecasting.

METHODS

An ecological study was conducted in the Special Region of Yogyakarta, Indonesia (January 2017-December 2022) using monthly panel data from 78 sub-districts. Secondary data sources included dengue surveillance (Health Office), meteorological data (NASA POWER), sociodemographic data (BPS-Statistics Indonesia), and land use data (Sentinel-2, ESRI). Predictors included rainfall, temperature, humidity, wind speed, atmospheric pressure, population density, and land use patterns. Data analysis was performed using R-INLA, with model performance assessed using Deviance Information Criterion (DIC), Watanabe-Akaike Information Criterion (WAIC), marginal log-likelihood, Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE).

RESULTS

The INLA-based Bayesian model effectively captured spatial and temporal dengue dynamics. Key predictors included rainfall lag 1 and 2 (mean = 0.001), temperature (mean = 0.151, CI: 0.090-0.210), humidity (mean = 0.056, CI: 0.040-0.073), built area (mean = 0.001), and water area (mean = 0.008, CI: 0.005-0.011). Spatial clustering (BYM model, precision = 2163.53) indicated that dengue cases were concentrated in specific areas. The RW2 model (precision = 49.11) confirmed seasonal trends, highlighting climate's role in disease transmission. Model evaluation metrics (DIC = 15017.88, WAIC = 15294.54, log-likelihood = -7845.857) demonstrated good predictive performance. Furthermore, the model's accuracy was assessed using MAE and RMSE values, where MAE = 1.77 indicates an average prediction error of 1-2 cases, while RMSE = 2.97 suggests the presence of occasional larger discrepancies. The RMSE's higher value relative to MAE highlights instances where prediction errors were more significant, as RMSE is more sensitive to large deviations.

CONCLUSIONS

The INLA-based spatio-temporal model is an effective tool for dengue prediction, offering valuable insights for early warning systems and targeted vector control strategies, thereby improving disease prevention and response efforts.

Forecasting dengue in Bangladesh using meteorological variables with a novel feature selection approach

Dengue, a mosquito-borne viral disease, continues to pose severe risks to public health and economic stability in tropical and subtropical regions, particularly in developing nations like Bangladesh. The necessity for advanced forecasting mechanisms has never been more critical to enhance the effectiveness of vector control strategies and resource allocations. This study formulates a dynamic data pipeline to forecast dengue incidence based on 13 meteorological variables using a suite of state-of-the-art machine learning models and custom features engineering, achieving an accuracy of 84.02%, marking a substantial improvement over existing studies. A novel wrapper feature selection algorithm employing a custom objective function is proposed in this study, which significantly improves model accuracy by 12.63% and reduces the mean absolute percentage error by 70.82%. The custom objective function's output can be transformed to quantify the contribution of each variable to the target variable's variability, providing deeper insights into the workings of black box models. The study concludes that relative humidity is redundant in predicting dengue infection, while meteorological factors exhibit more significant short-term impacts compared to long-term and immediate impacts. Sunshine (hours) emerges as the meteorological factor with the most immediate impact, whereas precipitation is the most impactful predictor over both short-term (8-month lag) and long-term (26-30-month lag) periods. Forecasts for 2024 using the best-performing model predict a rise in dengue cases starting in May, peaking at 24,000 cases per month by August and persisting at high levels through October before declining to half by year-end. These findings offer critical insights into temporal climate effects on dengue transmission, aiding the development of effective forecasting systems.

Spatial model of Dengue Hemorrhagic Fever (DHF) risk: scoping review

BACKGROUND

Creating a spatial model of dengue fever risk is challenging duet to many interrelated factors that could affect dengue. Therefore, it is crucial to understand how these critical factors interact and to create reliable predictive models that can be used to mitigate and control the spread of dengue.

METHODS

This scoping review aims to provide a comprehensive overview of the important predictors, and spatial modelling tools capable of producing Dengue Haemorrhagic Fever (DHF) risk maps. We conducted a methodical exploration utilizing diverse sources, i.e., PubMed, Scopus, Science Direct, and Google Scholar. The following data were extracted from articles published between January 2011 to August 2022: country, region, administrative level, type of scale, spatial model, dengue data use, and categories of predictors. Applying the eligibility criteria, 45 out of 1,349 articles were selected.

RESULTS

A variety of models and techniques were used to identify DHF risk areas with an arrangement of various multiple-criteria decision-making, statistical, and machine learning technique. We found that there was no pattern of predictor use associated with particular approaches. Instead, a wide range of predictors was used to create the DHF risk maps. These predictors may include climatology factors (e.g., temperature, rainfall, humidity), epidemiological factors (population, demographics, socio-economic, previous DHF cases), environmental factors (land-use, elevation), and relevant factors.

CONCLUSIONS

DHF risk spatial models are useful tools for detecting high-risk locations and driving proactive public health initiatives. Relying on geographical and environmental elements, these models ignored the impact of human behaviour and social dynamics. To improve the prediction accuracy, there is a need for a more comprehensive approach to understand DHF transmission dynamics.

Abdul Hamid AA, Zainuddin NH, Aruchunan E, Abdullah MT. Enhancing COVID-19 Classification Accuracy with a Hybrid SVM-LR Model

Support ector achine (SVM) is a newer machine learning algorithm for classification, while logistic regression (LR) is an older statistical classification method. Despite the numerous studies contrasting SVM and LR, new improvements such as bagging and ensemble have been applied to them since these comparisons were made. This study proposes a new hybrid model based on SVM and LR for predicting small events per variable (EPV). The performance of the hybrid, SVM, and LR models with different EPV values was evaluated using COVID-19 data from December 2019 to May 2020 provided by the WHO. The study found that the hybrid model had better classification performance than SVM and LR in terms of accuracy, mean squared error (MSE), and root mean squared error (RMSE) for different EPV values. This hybrid model is particularly important for medical authorities and practitioners working in the face of future pandemics.

Dengue in Myanmar: Spatiotemporal epidemiology, association with climate and short-term prediction

Dengue is a major public health problem in Myanmar. The country aims to reduce morbidity by 50% and mortality by 90% by 2025 based on 2015 data. To support efforts to reach these goals it is important to have a detailed picture of the epidemiology of dengue, its relationship to meteorological factors and ideally to predict ahead of time numbers of cases to plan resource allocations and control efforts. Health facility-level data on numbers of dengue cases from 2012 to 2017 were obtained from the Vector Borne Disease Control Unit, Department of Public Health, Myanmar. A detailed analysis of routine dengue and dengue hemorrhagic fever (DHF) incidence was conducted to examine the spatial and temporal epidemiology. Incidence was compared to climate data over the same period. Dengue was found to be widespread across the country with an increase in spatial extent over time. The temporal pattern of dengue cases and fatalities was episodic with annual outbreaks and no clear longitudinal trend. There were 127,912 reported cases and 632 deaths from 2012 and 2017 with peaks in 2013, 2015 and 2017. The case fatality rate was around 0.5% throughout. The peak season of dengue cases was from May to August in the wet season but in 2014 peak dengue season continued until November. The strength of correlation of dengue incidence with different climate factors (total rainfall, maximum, mean and minimum temperature and absolute humidity) varied between different States and Regions. Monthly incidence was forecasted 1 month ahead using the Auto Regressive Integrated Moving Average (ARIMA) method at country and subnational levels. With further development and validation, this may be a simple way to quickly generate short-term predictions at subnational scales with sufficient certainty to use for intervention planning.

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.

A Systematic Review of Applications of Machine Learning and Other Soft Computing Techniques for the Diagnosis of Tropical Diseases

This systematic literature aims to identify soft computing techniques currently utilized in diagnosing tropical febrile diseases and explore the data characteristics and features used for diagnoses, algorithm accuracy, and the limitations of current studies. The goal of this study is therefore centralized around determining the extent to which soft computing techniques have positively impacted the quality of physician care and their effectiveness in tropical disease diagnosis. The study has used PRISMA guidelines to identify paper selection and inclusion/exclusion criteria. It was determined that the highest frequency of articles utilized ensemble techniques for classification, prediction, analysis, diagnosis, etc., over single machine learning techniques, followed by neural networks. The results identified dengue fever as the most studied disease, followed by malaria and tuberculosis. It was also revealed that accuracy was the most common metric utilized to evaluate the predictive capability of a classification mode. The information presented within these studies benefits frontline healthcare workers who could depend on soft computing techniques for accurate diagnoses of tropical diseases. Although our research shows an increasing interest in using machine learning techniques for diagnosing tropical diseases, there still needs to be more studies. Hence, recommendations and directions for future research are proposed.

Development and evaluation of a visible reverse transcription-loop-mediated isothermal amplification (RT-LAMP) for the detection of Asian lineage ZIKV in field-caught mosquitoes

The Zika virus (ZIKV) infection is an emerging and re-emerging arbovirus infection that is transmitted to humans through the bite of infected mosquitoes. Early detection of ZIKV in mosquitoes is one of the prerequisite approaches for tracking the spread of the virus. Therefore, this study aims to develop and validate a visual reverse transcription-loop-mediated isothermal amplification (RT-LAMP) method called ZIKV-RT-LAMP, for detecting ZIKV in field collected mosquito samples from Thailand. A single-tube ZIKV-RT-LAMP assay was developed to detect Asian lineage ZIKV RNA. The detection limit and cross-reactivity of ZIKV were investigated. The hemi-nested RT-PCR (hn-RT-PCR) and the colorimetric LAMP kit (cLAMP kit) were performed as reference assays. The detection limit of the ZIKV-RT-LAMP assay was 10^-6 ffu/ml or pfu/ml, making it highly specific and 100 times more sensitive than the hn-RT-PCR and cLAMP kits. The ZIKV-RT-LAMP assay detected the Asian lineage of ZIKV RNA without cross-reactivity with other arthropod-borne viruses. The sensitivity and specificity of the ZIKV-RT-LAMP assay were 92.31% and 100%, respectively. The ZIKV-RT-LAMP is a simple, rapid, and inexpensive method for detecting ZIKV in field-caught mosquitos. In the future, extensive surveys of field-caught mosquito populations should be conducted. Early detection of ZIKV in field-caught mosquitoes provides for prompt and effective implementation of mosquito control strategies in endemic areas.

Forecasting the incidence of dengue in Bangladesh-Application of time series model

Background

Dengue is an alarming public health concern in terms of its preventive and curative measures among people in Bangladesh; moreover, its sudden outbreak created a lot of suffering among people in 2018. Considering the greater burden of disease in larger epidemic years and the difficulty in understanding current and future needs, it is highly needed to address early warning systems to control epidemics from the earliest.

Objective

The study objective was to select the most appropriate model for dengue incidence and using the selected model, the authors forecast the future dengue outbreak in Bangladesh.

Methods and Materials

This study considered a secondary data set of monthly dengue occurrences over the period of January 2008 to January 2020. Initially, the authors found the suitable model from Autoregressive Integrated Moving Average (ARIMA), Error, Trend, Seasonal (ETS) and Trigonometric seasonality, Box-Cox transformation, ARMA errors, Trend and Seasonal (TBATS) models with the help of selected model selection criteria and finally employing the selected model make forecasting of dengue incidences in Bangladesh.

Results

Among ARIMA, ETS, and TBATS models, the ARIMA model performs better than others. The Box-Jenkin's procedure is applicable here and it is found that the best-selected model to forecast the dengue outbreak in the context of Bangladesh is ARIMA (2,1,2).

Conclusion

Before establishing a comprehensive plan for future combating strategies, it is vital to understand the future scenario of dengue occurrence. With this in mind, the authors aimed to select an appropriate model that might predict dengue fever outbreaks in Bangladesh. The findings revealed that dengue fever is expected to become more frequent in the future. The authors believe that the study findings will be helpful to take early initiatives to combat future dengue outbreaks.

Temporal and Spatiotemporal Arboviruses Forecasting by Machine Learning: A Systematic Review

Arboviruses are a group of diseases that are transmitted by an arthropod vector. Since they are part of the Neglected Tropical Diseases that pose several public health challenges for countries around the world. The arboviruses' dynamics are governed by a combination of climatic, environmental, and human mobility factors. Arboviruses prediction models can be a support tool for decision-making by public health agents. In this study, we propose a systematic literature review to identify arboviruses prediction models, as well as models for their transmitter vector dynamics. To carry out this review, we searched reputable scientific bases such as IEE Xplore, PubMed, Science Direct, Springer Link, and Scopus. We search for studies published between the years 2015 and 2020, using a search string. A total of 429 articles were returned, however, after filtering by exclusion and inclusion criteria, 139 were included. Through this systematic review, it was possible to identify the challenges present in the construction of arboviruses prediction models, as well as the existing gap in the construction of spatiotemporal models.

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

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.

Dengue fever (DF) represents a significant health burden worldwide and 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. This study aimed to use deep learning models to develop a prediction model of DF rates in Vietnam using a wide range of climate factors as input variables to inform public health responses for outbreak prevention in the context of future climate change. The study found that LSTM-ATT outperformed competing models, scoring average places of 1.60 for RMSE-based ranking and 1.90 for MAE-based ranking. Notably, it was able to forecast DF incidence better than LSTM in 12 or 14 out of 20 provinces for MAE or RMSE, respectively. Moreover, LSTM-ATT was able to accurately predict DF incidence and outbreaks up to 3 months ahead, though performance dropped slightly compared to short-term forecasts. 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 climate features, and it demonstrates the usefulness of deep learning models for climate-based DF forecasting.

A multivariate multi-step LSTM forecasting model for tuberculosis incidence with model explanation in Liaoning Province, China

BACKGROUND

Tuberculosis (TB) is the respiratory infectious disease with the highest incidence in China. We aim to design a series of forecasting models and find the factors that affect the incidence of TB, thereby improving the accuracy of the incidence prediction.

RESULTS

In this paper, we developed a new interpretable prediction system based on the multivariate multi-step Long Short-Term Memory (LSTM) model and SHapley Additive exPlanation (SHAP) method. Four accuracy measures are introduced into the system: Root Mean Square Error, Mean Absolute Error, Mean Absolute Percentage Error, and symmetric Mean Absolute Percentage Error. The Autoregressive Integrated Moving Average (ARIMA) model and seasonal ARIMA model are established. The multi-step ARIMA-LSTM model is proposed for the first time to examine the performance of each model in the short, medium, and long term, respectively. Compared with the ARIMA model, each error of the multivariate 2-step LSTM model is reduced by 12.92%, 15.94%, 15.97%, and 14.81% in the short term. The 3-step ARIMA-LSTM model achieved excellent performance, with each error decreased to 15.19%, 33.14%, 36.79%, and 29.76% in the medium and long term. We provide the local and global explanation of the multivariate single-step LSTM model in the field of incidence prediction, pioneering.

CONCLUSIONS

The multivariate 2-step LSTM model is suitable for short-term prediction and obtained a similar performance as previous studies. The 3-step ARIMA-LSTM model is appropriate for medium-to-long-term prediction and outperforms these models. The SHAP results indicate that the five most crucial features are maximum temperature, average relative humidity, local financial budget, monthly sunshine percentage, and sunshine hours.

Dengue Early Warning System as Outbreak Prediction Tool: A Systematic Review

Early warning system (EWS) for vector-borne diseases is incredibly complex due to numerous factors originating from human, environmental, vector and the disease itself. Dengue EWS aims to collect data that leads to prompt decision-making processes that trigger disease intervention strategies to minimize the impact on a specific population. Dengue EWS may have a similar structural design, functions, and analytical approaches but different performance and ability to predict outbreaks. Hence, this review aims to summarise and discuss the evidence of different EWSs, their performance, and their ability to predict dengue outbreaks. A systematic literature search was performed of four primary databases: Scopus, Web of Science, Ovid MEDLINE, and EBSCOhost. Eligible articles were evaluated using a checklist for assessing the quality of the studies. A total of 17 studies were included in this systematic review. All EWS models demonstrated reasonably good predictive abilities to predict dengue outbreaks. However, the accuracy of their predictions varied greatly depending on the model used and the data quality. The reported sensitivity ranged from 50 to 100%, while specificity was 74 to 94.7%. A range between 70 to 96.3% was reported for prediction model accuracy and 43 to 86% for PPV. Overall, meteorological alarm indicators (temperatures and rainfall) were the most frequently used and displayed the best performing indicator. Other potential alarm indicators are entomology (female mosquito infection rate), epidemiology, population and socioeconomic factors. EWS is an essential tool to support district health managers and national health planners to mitigate or prevent disease outbreaks. This systematic review highlights the benefits of integrating several epidemiological tools focusing on incorporating climatic, environmental, epidemiological and socioeconomic factors to create an early warning system. The early warning system relies heavily on the country surveillance system. The lack of timely and high-quality data is critical for developing an effective EWS.

Epidemic prediction of dengue fever based on vector compartment model and Markov chain Monte Carlo method

Background

Dengue epidemics is affected by vector-human interactive dynamics. Infectious disease prevention and control emphasize the timing intervention at the right diffusion phase. In such a way, control measures can be cost-effective, and epidemic incidents can be controlled before devastated consequence occurs. However, timing relations between a measurable signal and the onset of the pandemic are complex to be discovered, and the typical lag period regression is difficult to capture in these complex relations. This study investigates the dynamic diffusion pattern of the disease in terms of a probability distribution. We estimate the parameters of an epidemic compartment model with the cross-infection of patients and mosquitoes in various infection cycles. We comprehensively study the incorporated meteorological and mosquito factors that may affect the epidemic of dengue fever to predict dengue fever epidemics.

Results

We develop a dual-parameter estimation algorithm for a composite model of the partial differential equations for vector-susceptible-infectious-recovered with exogeneity compartment model, Markov chain Montel Carlo method, and boundary element method to evaluate the epidemic periodicity under the effect of environmental factors of dengue fever, given the time series data of 2000–2016 from three cities with a population of 4.7 million. The established computer model of “energy accumulation-delayed diffusion-epidemics” is proven to be effective to predict the future trend of reported and unreported infected incidents. Our artificial intelligent algorithm can inform the authority to cease the larvae at the highest vector infection time. We find that the estimated dengue report rate is about 20%, which is close to the number of official announcements, and the percentage of infected vectors increases exponentially yearly. We suggest that the executive authorities should seriously consider the accumulated effect among infected populations. This established epidemic prediction model of dengue fever can be used to simulate and evaluate the best time to prevent and control dengue fever.

Conclusions

Given our developed model, government epidemic prevention teams can apply this platform before they physically carry out the prevention work. The optimal suggestions from these models can be promptly accommodated when real-time data have been continuously corrected from clinics and related agents.

Low yield but high levels of multidrug resistance in urinary tract infections in a tertiary hospital, Nepal

SETTING

There are concerns about the occurrence of multidrug resistance (MDR) in patients with urine tract infections (UTI) in Nepal.

OBJECTIVE

To determine culture positivity, trends in MDR among Escherichia coli and Klebsiella pneumoniae infections and seasonal changes in culture-positive UTI specimens isolated from 2014 to 2018 at the B P Koirala Institute of Health Sciences, Dharan, Eastern Nepal.

DESIGN

This was a cross-sectional study using secondary laboratory data.

RESULTS

Among 116,417 urine samples tested, 19,671 (16.9%) were culture-positive, with an increasing trend in the number of samples tested and culture positivity. E. coli was the most common bacteria (54.3%), followed by K. pneumoniae (8.8%). Among E. coli and K. pneumoniae isolates, MDR was found in respectively 42.5% and 36.0%. MDR was higher in males and people aged >55 years, but showed a decreasing trend over the years. The numbers of isolates increased over the years, with a peak always observed from July to August.

CONCLUSION

Low culture positivity is worrying and requires further work into improving diagnostic protocols. Decreasing trends in MDR are a welcome sign. Information on seasonal changes that peak in July-August can help laboratories better prepare for this time with adequate buffer stocks to ensure culture and antibiotic susceptibility testing.

Enhancing fine-grained intra-urban dengue forecasting by integrating spatial interactions of human movements between urban regions

Background

As a mosquito-borne infectious disease, dengue fever (DF) has spread through tropical and subtropical regions worldwide in recent decades. Dengue forecasting is essential for enhancing the effectiveness of preventive measures. Current studies have been primarily conducted at national, sub-national, and city levels, while an intra-urban dengue forecasting at a fine spatial resolution still remains a challenging feat. As viruses spread rapidly because of a highly dynamic population flow, integrating spatial interactions of human movements between regions would be potentially beneficial for intra-urban dengue forecasting.

Methodology

In this study, a new framework for enhancing intra-urban dengue forecasting was developed by integrating the spatial interactions between urban regions. First, a graph-embedding technique called Node2Vec was employed to learn the embeddings (in the form of an N-dimensional real-valued vector) of the regions from their population flow network. As strongly interacting regions would have more similar embeddings, the embeddings can serve as “interaction features.” Then, the interaction features were combined with those commonly used features (e.g., temperature, rainfall, and population) to enhance the supervised learning–based dengue forecasting models at a fine-grained intra-urban scale.

Results

The performance of forecasting models (i.e., SVM, LASSO, and ANN) integrated with and without interaction features was tested and compared on township-level dengue forecasting in Guangzhou, the most threatened sub-tropical city in China. Results showed that models using both common and interaction features can achieve better performance than that using common features alone.

Conclusions

The proposed approach for incorporating spatial interactions of human movements using graph-embedding technique is effective, which can help enhance fine-grained intra-urban dengue forecasting.

Dengue fever, a mosquito-borne infectious disease, has become a serious public health problem in many tropical and subtropical regions worldwide, such as Southeast Asian countries and the Guangdong Province in China. In the absence of an effective vaccine at present, disease surveillance and mosquito control remain the primary means of controlling the spread of the disease. At an intra-urban setting, it is important to predict the spatial distribution of future patients, which can help government agencies to establish precise and targeted prevention measures beforehand. Considering the fast virus spread within a city because of a highly dynamic population flow, we proposed a novel approach to enhancing fine-grained intra-urban dengue forecasting by integrating spatial interactions of human movements between urban regions. First, using a graph-embedding model called Node2Vec, the embeddings of the regions were learned from their population interaction network so that strongly interacted regions would have more similar embeddings. Secondly, serving as interaction features, the embeddings were combined with the commonly used features as inputs of the forecasting models. The experimental results indicated that the performance of the models can be improved by incorporating the interaction features, confirming the effectiveness of our proposed strategy in enhancing fine-grained intra-urban dengue forecasting.

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.

The time series seasonal patterns of dengue fever and associated weather variables in Bangkok (2003-2017)

Background

In Thailand, dengue fever is one of the most well-known public health problems. The objective of this study was to examine the epidemiology of dengue and determine the seasonal pattern of dengue and its associate to climate factors in Bangkok, Thailand, from 2003 to 2017.

Methods

The dengue cases in Bangkok were collected monthly during the study period. The time-series data were extracted into the trend, seasonal, and random components using the seasonal decomposition procedure based on loess. The Spearman correlation analysis and artificial neuron network (ANN) were used to determine the association between climate variables (humidity, temperature, and rainfall) and dengue cases in Bangkok.

Results

The seasonal-decomposition procedure showed that the seasonal component was weaker than the trend component for dengue cases during the study period. The Spearman correlation analysis showed that rainfall and humidity played a role in dengue transmission with correlation efficiency equal to 0.396 and 0.388, respectively. ANN showed that precipitation was the most crucial factor. The time series multivariate Poisson regression model revealed that increasing 1% of rainfall corresponded to an increase of 3.3% in the dengue cases in Bangkok. There were three models employed to forecast the dengue case, multivariate Poisson regression, ANN, and ARIMA. Each model displayed different accuracy, and multivariate Poisson regression was the most accurate approach in this study.

Conclusion

This work demonstrates the significance of weather in dengue transmission in Bangkok and compares the accuracy of the different mathematical approaches to predict the dengue case. A single model may insufficient to forecast precisely a dengue outbreak, and climate factor may not only indicator of dengue transmissibility.

A dynamic neural network model for predicting risk of Zika in real time

BACKGROUND

In 2015, the Zika virus spread from Brazil throughout the Americas, posing an unprecedented challenge to the public health community. During the epidemic, international public health officials lacked reliable predictions of the outbreak's expected geographic scale and prevalence of cases, and were therefore unable to plan and allocate surveillance resources in a timely and effective manner.

METHODS

In this work, we present a dynamic neural network model to predict the geographic spread of outbreaks in real time. The modeling framework is flexible in three main dimensions (i) selection of the chosen risk indicator, i.e., case counts or incidence rate; (ii) risk classification scheme, which defines the high-risk group based on a relative or absolute threshold; and (iii) prediction forecast window (1 up to 12 weeks). The proposed model can be applied dynamically throughout the course of an outbreak to identify the regions expected to be at greatest risk in the future.

RESULTS

The model is applied to the recent Zika epidemic in the Americas at a weekly temporal resolution and country spatial resolution, using epidemiological data, passenger air travel volumes, and vector habitat suitability, socioeconomic, and population data for all affected countries and territories in the Americas. The model performance is quantitatively evaluated based on the predictive accuracy of the model. We show that the model can accurately predict the geographic expansion of Zika in the Americas with the overall average accuracy remaining above 85% even for prediction windows of up to 12 weeks.

CONCLUSIONS

Sensitivity analysis illustrated the model performance to be robust across a range of features. Critically, the model performed consistently well at various stages throughout the course of the outbreak, indicating its potential value at any time during an epidemic. The predictive capability was superior for shorter forecast windows and geographically isolated locations that are predominantly connected via air travel. The highly flexible nature of the proposed modeling framework enables policy makers to develop and plan vector control programs and case surveillance strategies which can be tailored to a range of objectives and resource constraints.

Distribution of and associated factors for dengue burden in the state of Odisha, India during 2010-2016

This study is aimed to estimate the epidemiological burden of dengue in Odisha, India using the disability adjusted life year (DALY) methods and to explore the associated factors in the year 2010-2016. During the period of 2010-2016, 27 772 cases (68.4% male) were reported in the state. Mean age (years) of male and female was 31.63 and 33.82, respectively. Mean district wise disability adjusted life years (DALY) per 100 000 people was higher in the year 2016 (0.45) and mean DALY lost per person was highest in the year 2015 (34.90 years). Adjusted regression model indicates, every unit increase in humidity and population density increases DALY by 1.05 and 1.02 units respectively. Whereas, unit change in sex ratio (females per 1000 males) and forest coverage increases the DALY by 0.98 units. Our results indicate geographical variation of DALY in Odisha, which is associated with population density, humidity and forest cover. Discrepancies identified between standard incidence and DALY maps suggests, latter can be used to present disease burden more effectively. More prevalence among young males suggests the need of strengthening the targeted prevention and control measures.

Severe Dengue Prognosis Using Human Genome Data and Machine Learning

Dengue has become one of the most important worldwide arthropod-borne diseases. Dengue phenotypes are based on laboratorial and clinical exams, which are known to be inaccurate.

OBJECTIVE

We present a machine learning approach for the prediction of dengue fever severity based solely on human genome data.

METHODS

One hundred and two Brazilian dengue patients and controls were genotyped for 322 innate immunity single nucleotide polymorphisms (SNPs). Our model uses a support vector machine algorithm to find the optimal loci classification subset and then an artificial neural network (ANN) is used to classify patients into dengue fever or severe dengue.

RESULTS

The ANN trained on 13 key immune SNPs selected under dominant or recessive models produced median values of accuracy greater than 86%, and sensitivity and specificity over 98% and 51%, respectively.

CONCLUSION

The proposed classification method, using only genome markers, can be used to identify individuals at high risk for developing the severe dengue phenotype even in uninfected conditions.

SIGNIFICANCE

Our results suggest that the genetic context is a key element in phenotype definition in dengue. The methodology proposed here is extendable to other Mendelian based and genetically influenced diseases.

A Novel Machine Learning Algorithm to Automatically Predict Visual Outcomes in Intravitreal Ranibizumab-Treated Patients with Diabetic Macular Edema

Purpose: Artificial neural networks (ANNs) are one type of artificial intelligence. Here, we use an ANN-based machine learning algorithm to automatically predict visual outcomes after ranibizumab treatment in diabetic macular edema. Methods: Patient data were used to optimize ANNs for regression calculation. The target was established as the final visual acuity at 52, 78, or 104 weeks. The input baseline variables were sex, age, diabetes type or condition, systemic diseases, eye status and treatment time tables. Three groups were randomly devised to build, test and demonstrate the accuracy of the algorithms. Results: At 52, 78 and 104 weeks, 512, 483 and 464 eyes were included, respectively. For the training group, testing group and validation group, the respective correlation coefficients were 0.75, 0.77 and 0.70 (52 weeks); 0.79, 0.80 and 0.55 (78 weeks); and 0.83, 0.47 and 0.81 (104 weeks), while the mean standard errors of final visual acuity were 6.50, 6.11 and 6.40 (52 weeks); 5.91, 5.83 and 7.59; (78 weeks); and 5.39, 8.70 and 6.81 (104 weeks). Conclusions: Machine learning had good correlation coefficients for predicating prognosis with ranibizumab with just baseline characteristics. These models could be the useful clinical tools for prediction of success of the treatments.

The Diagnosis of Dengue Disease

This article describes how Dengue fever is a fatal and hazardous disease resulting from the bite of several species of the female mosquito (principally, Aedesaegypti). Symptoms of the dengue fever mimic those of a number of other infectious and/or mosquito-borne tropical diseases such as Viral flu, Chikungunya, and Zika fever. Yet, with dengue fever, human life can be more at risk due to severe depletion of blood platelets. Thus, early detection of dengue disease can ensure saving lives; furthermore, it can help in making a preventive move before the disease progresses to epidemic proportion. Hence, the target of this article is to propose a model for an early detection and precise diagnosis of dengue disease. In this article, three prevalent machine learning methodologies, including, Artificial Neural Network (ANN), Decision Tree (DT) and Naive Bayes (NB) are evaluated for designing a diagnostic model. The performance of these models is assessed utilizing available dengue datasets. Results comparing and contrasting performance of diagnostic models utilizing accuracy, sensitivity, specificity and error rate parameters showed that ANN-based diagnostic model appears to yield better performance measures over both the DT and NB models.

The role of UV radiation and vitamin D in the seasonality and outcomes of infectious disease

The seasonality of infectious disease outbreaks suggests that environmental conditions have a significant effect on disease risk. One of the major environmental factors that can affect this is solar radiation, primarily acting through ultraviolet radiation (UVR), and its subsequent control of vitamin D production. Here we show how UVR and vitamin D, which are modified by latitude and season, can affect host and pathogen fitness and relate them to the outcomes of bacterial, viral and vector-borne infections. We conducted a thorough comparison of the molecular and cellular mechanisms of action of UVR and vitamin D on pathogen fitness and host immunity and related these to the effects observed in animal models and clinical trials to understand their independent and complementary effects on infectious disease outcome. UVR and vitamin D share common pathways of innate immune activation primarily via antimicrobial peptide production, and adaptive immune suppression. Whilst UVR can induce vitamin D-independent effects in the skin, such as the generation of photoproducts activating interferon signaling, vitamin D has a larger systemic effect due to its autocrine and paracrine modulation of cellular responses in a range of tissues. However, the seasonal patterns in infectious disease prevalence are not solely driven by variation in UVR and vitamin D levels across latitudes. Vector-borne pathogens show a strong seasonality of infection correlated to climatic conditions favoring their replication. Conversely, pathogens, such as influenza A virus, Mycobacterium tuberculosis and human immunodeficiency virus type 1, have strong evidence to support their interaction with vitamin D. Thus, UVR has both vitamin D-dependent and independent effects on infectious diseases; these effects vary depending on the pathogen of interest and the effects can be complementary or antagonistic.

Time series model for forecasting the number of new admission inpatients

Background

Hospital crowding is a rising problem, effective predicting and detecting managment can helpful to reduce crowding. Our team has successfully proposed a hybrid model combining both the autoregressive integrated moving average (ARIMA) and the nonlinear autoregressive neural network (NARNN) models in the schistosomiasis and hand, foot, and mouth disease forecasting study. In this paper, our aim is to explore the application of the hybrid ARIMA-NARNN model to track the trends of the new admission inpatients, which provides a methodological basis for reducing crowding.

Methods

We used the single seasonal ARIMA (SARIMA), NARNN and the hybrid SARIMA-NARNN model to fit and forecast the monthly and daily number of new admission inpatients. The root mean square error (RMSE), mean absolute error (MAE) and mean absolute percentage error (MAPE) were used to compare the forecasting performance among the three models. The modeling time range of monthly data included was from January 2010 to June 2016, July to October 2016 as the corresponding testing data set. The daily modeling data set was from January 4 to September 4, 2016, while the testing time range included was from September 5 to October 2, 2016.

Results

For the monthly data, the modeling RMSE and the testing RMSE, MAE and MAPE of SARIMA-NARNN model were less than those obtained from the single SARIMA or NARNN model, but the MAE and MAPE of modeling performance of SARIMA-NARNN model did not improve. For the daily data, all RMSE, MAE and MAPE of NARNN model were the lowest both in modeling stage and testing stage.

Conclusions

Hybrid model does not necessarily outperform its constituents’ performances. It is worth attempting to explore the reliable model to forecast the number of new admission inpatients from different data.

Electronic supplementary material

The online version of this article (10.1186/s12911-018-0616-8) contains supplementary material, which is available to authorized users.

Forecasting zoonotic cutaneous leishmaniasis using meteorological factors in eastern Fars province, Iran: a SARIMA analysis

OBJECTIVES

To predict the occurrence of zoonotic cutaneous leishmaniasis (ZCL) and evaluate the effect of climatic variables on disease incidence in the east of Fars province, Iran using the Seasonal Autoregressive Integrated Moving Average (SARIMA) model.

METHODS

The Box-Jenkins approach was applied to fit the SARIMA model for ZCL incidence from 2004 to 2015. Then the model was used to predict the number of ZCL cases for the year 2016. Finally, we assessed the relation of meteorological variables (rainfall, rainy days, temperature, hours of sunshine and relative humidity) with ZCL incidence.

RESULTS

SARIMA(2,0,0) (2,1,0)12 was the preferred model for predicting ZCL incidence in the east of Fars province (validation Root Mean Square Error, RMSE = 0.27). It showed that ZCL incidence in a given month can be estimated by the number of cases occurring 1 and 2 months, as well as 12 and 24 months earlier. The predictive power of SARIMA models was improved by the inclusion of rainfall at a lag of 2 months (β = -0.02), rainy days at a lag of 2 months (β = -0.09) and relative humidity at a lag of 8 months (β = 0.13) as external regressors (P-values < 0.05). The latter was the best climatic variable for predicting ZCL cases (validation RMSE = 0.26).

CONCLUSIONS

Time series models can be useful tools to predict the trend of ZCL in Fars province, Iran; thus, they can be used in the planning of public health programmes. Introducing meteorological variables into the models may improve their precision.

Decoding and modelling of time series count data using Poisson hidden Markov model and Markov ordinal logistic regression models

Hidden Markov models are stochastic models in which the observations are assumed to follow a mixture distribution, but the parameters of the components are governed by a Markov chain which is unobservable. The issues related to the estimation of Poisson-hidden Markov models in which the observations are coming from mixture of Poisson distributions and the parameters of the component Poisson distributions are governed by an m-state Markov chain with an unknown transition probability matrix are explained here. These methods were applied to the data on Vibrio cholerae counts reported every month for 11-year span at Christian Medical College, Vellore, India. Using Viterbi algorithm, the best estimate of the state sequence was obtained and hence the transition probability matrix. The mean passage time between the states were estimated. The 95% confidence interval for the mean passage time was estimated via Monte Carlo simulation. The three hidden states of the estimated Markov chain are labelled as 'Low', 'Moderate' and 'High' with the mean counts of 1.4, 6.6 and 20.2 and the estimated average duration of stay of 3, 3 and 4 months, respectively. Environmental risk factors were studied using Markov ordinal logistic regression analysis. No significant association was found between disease severity levels and climate components.

Relevance of transportation to correlations among criticality, physical means of propagation, and distribution of dengue fever cases in the state of Bahia

Dengue infection is a public health problem with a complex distribution. The physical means of propagation and the dynamics of diffusion of the disease between municipalities need to be analysed to direct efficient public policies to prevent dengue infection. The present study presents correlations of occurrences of reported cases of dengue infection among municipalities, self-organized criticality (SOC), and transportation between areas, identifying the municipalities that play an important role in the diffusion of dengue across the state of Bahia, Brazil. The significant correlation found between the correlation network and the SOC demonstrates that the pattern of intramunicipal diffusion of dengue is coupled to the pattern of synchronisation between the municipalities. Transportation emerges as influential in the dynamics of diffusion of epidemics by acting on the aforementioned variables.

An economic evaluation of vector control in the age of a dengue vaccine

INTRODUCTION

Dengue is a rapidly emerging vector-borne Neglected Tropical Disease, with a 30-fold increase in the number of cases reported since 1960. The economic cost of the illness is measured in the billions of dollars annually. Environmental change and unplanned urbanization are conspiring to raise the health and economic cost even further beyond the reach of health systems and households. The health-sector response has depended in large part on control of the Aedes aegypti and Ae. albopictus (mosquito) vectors. The cost-effectiveness of the first-ever dengue vaccine remains to be evaluated in the field. In this paper, we examine how it might affect the cost-effectiveness of sustained vector control.

METHODS

We employ a dynamic Markov model of the effects of vector control on dengue in both vectors and humans over a 15-year period, in six countries: Brazil, Columbia, Malaysia, Mexico, the Philippines, and Thailand. We evaluate the cost (direct medical costs and control programme costs) and cost-effectiveness of sustained vector control, outbreak response and/or medical case management, in the presence of a (hypothetical) highly targeted and low cost immunization strategy using a (non-hypothetical) medium-efficacy vaccine.

RESULTS

Sustained vector control using existing technologies would cost little more than outbreak response, given the associated costs of medical case management. If sustained use of existing or upcoming technologies (of similar price) reduce vector populations by 70-90%, the cost per disability-adjusted life year averted is 2013 US$ 679-1331 (best estimates) relative to no intervention. Sustained vector control could be highly cost-effective even with less effective technologies (50-70% reduction in vector populations) and in the presence of a highly targeted and low cost immunization strategy using a medium-efficacy vaccine.

DISCUSSION

Economic evaluation of the first-ever dengue vaccine is ongoing. However, even under very optimistic assumptions about a highly targeted and low cost immunization strategy, our results suggest that sustained vector control will continue to play an important role in mitigating the impact of environmental change and urbanization on human health. If additional benefits for the control of other Aedes borne diseases, such as Chikungunya, yellow fever and Zika fever are taken into account, the investment case is even stronger. High-burden endemic countries should proceed to map populations to be covered by sustained vector control.

Effect of Rainfall for the Dynamical Transmission Model of the Dengue Disease in Thailand

The SEIR (Susceptible-Exposed-Infected-Recovered) model is used to describe the transmission of dengue virus. The main contribution is determining the role of the rainfall in Thailand in the model. The transmission of dengue disease is assumed to depend on the nature of the rainfall in Thailand. We analyze the dynamic transmission of dengue disease. The stability of the solution of the model is analyzed. It is investigated by using the Routh-Hurwitz criteria. We find two equilibrium states: a disease-free state and an endemic equilibrium state. The basic reproductive number (R₀) is obtained, which indicates the stability of each equilibrium state. Numerical results taking into account the rainfall are obtained and they are seen to correspond to the analytical results.

Neural network prediction of severe lower intestinal bleeding and the need for surgical intervention

BACKGROUND

The prognosis for patients with severe acute lower intestinal bleeding (ALIB) may be assessed by complex artificial neural networks (ANNs) or user-friendly regression-based models. Comparisons between these modalities are limited, and predicting the need for surgical intervention remains elusive. We hypothesized that ANNs would outperform the Strate rule to predict severe bleeding and would also predict the need for surgical intervention.

METHODS

We performed a 4-y retrospective analysis of 147 adult patients who underwent endoscopy, angiography, or surgery for ALIB. Baseline characteristics, Strate risk factors, management parameters, and outcomes were analyzed. The primary outcomes were severe bleeding and surgical intervention. ANNs were created in SPSS. Models were compared by area under the receiver operating characteristic curve (AUROC) with 95% confidence intervals.

RESULTS

The number of Strate risk factors for each patient correlated significantly with the outcome of severe bleeding (r = 0.29, P < 0.001). However, the Strate model was less accurate than an ANN (AUROC 0.66 [0.57-0.75] versus 0.98 [0.95-1.00], respectively) which incorporated six variables present on admission: hemoglobin, systolic blood pressure, outpatient prescription for Aspirin 325 mg daily, Charlson comorbidity index, base deficit ≥5 mEq/L, and international normalized ratio ≥1.5. A similar ANN including hemoglobin nadir and the occurrence of a 20% decrease in hematocrit was effective in predicting the need for surgery (AUROC 0.95 [0.90-1.00]).

CONCLUSIONS

The Strate prediction rule effectively stratified risk for severe ALIB, but was less accurate than an ANN. A separate ANN accurately predicted the need for surgery by combining risk factors for severe bleeding with parameters quantifying blood loss. Optimal prognostication may be achieved by integrating pragmatic regression-based calculators for quick decisions at the bedside and highly accurate ANNs when time and resources permit.