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

Modelling techniques in cholera epidemiology: A systematic and critical review

Diverse modelling techniques in cholera epidemiology have been developed and used to (1) study its transmission dynamics, (2) predict and manage cholera outbreaks, and (3) assess the impact of various control and mitigation measures. In this study, we carry out a critical and systematic review of various approaches used for modelling the dynamics of cholera. Also, we discuss the strengths and weaknesses of each modelling approach. A systematic search of articles was conducted in Google Scholar, PubMed, Science Direct, and Taylor & Francis. Eligible studies were those concerned with the dynamics of cholera excluding studies focused on models for cholera transmission in animals, socio-economic factors, and genetic & molecular related studies. A total of 476 peer-reviewed articles met the inclusion criteria, with about 40% (32%) of the studies carried out in Asia (Africa). About 52%, 21%, and 9%, of the studies, were based on compartmental (e.g., SIRB), statistical (time series and regression), and spatial (spatiotemporal clustering) models, respectively, while the rest of the analysed studies used other modelling approaches such as network, machine learning and artificial intelligence, Bayesian, and agent-based approaches. Cholera modelling studies that incorporate vector/housefly transmission of the pathogen are scarce and a small portion of researchers (3.99%) considers the estimation of key epidemiological parameters. Vaccination only platform was utilized as a control measure in more than half (58%) of the studies. Research productivity in cholera epidemiological modelling studies have increased in recent years, but authors used diverse range of models. Future models should consider incorporating vector/housefly transmission of the pathogen and on the estimation of key epidemiological parameters for the transmission of cholera dynamics.

How can machine learning predict cholera: insights from experiments and design science for action research

Cholera is a leading cause of mortality in Nigeria. The two most significant predictors of cholera are a lack of access to clean water and poor sanitary conditions. Other factors such as natural disasters, illiteracy, and internal conflicts that drive people to seek sanctuary in refugee camps may contribute to the spread of cholera in Nigeria. The aim of this research is to develop a cholera outbreak risk prediction (CORP) model using machine learning tools and data science. In this study, we developed a CORP model using design science perspectives and machine learning to detect cholera outbreaks in Nigeria. Nonnegative matrix factorization (NMF) was used for dimensionality reduction, and synthetic minority oversampling technique (SMOTE) was used for data balancing. Outliers were detected using density-based spatial clustering of applications with noise (DBSCAN) were removed improving the overall performance of the model, and the extreme-gradient boost algorithm was used for prediction. The findings revealed that the CORP model outcomes resulted in the best accuracy of 99.62%, Matthews's correlation coefficient of 0.976, and area under the curve of 99.2%, which were improved compared with the previous findings. The developed model can be helpful to healthcare providers in predicting possible cholera outbreaks.

The concern about main crises such as the Covid-19 pandemic, the war in Ukraine, and climate change's impact on energy-saving behavior

The number of crises experienced around the world forces people to reconsider and reassess various aspects of their lives. The energy crisis caused by the war in Ukraine and uncontrolled climate change revealed the importance of energy-saving behavior. Thus, the aim of this paper is to analyze the concerns about current crises such as the Covid-19 pandemic, the war in Ukraine, and climate change's impact on energy-saving behavior and changes in environmental concern. Referring to the survey conducted in Lithuania in 2022, where 1000 respondents participated, the results revealed that the war in Ukraine was the most concerning problem. The level of climate change concern was slightly lower. Meanwhile, the Covid-19 pandemic was the least important problem in Lithuania in 2022. Furthermore, respondents stated that the Covid-19 pandemic contributed to the changes in environmental concern and energy-saving actions more than the war in Ukraine did. Meanwhile, the Generalized Linear Model results revealed that only the war in Ukraine positively and significantly influenced energy-saving behavior. The Covid-19 pandemic concern negatively affected energy-saving behavior, while the climate change concern factor affected it indirectly, as the interaction of attitudes toward energy consumption. Thus, this study revealed the main aspect of and how to encourage energy-saving behavior in the context of the main current crises.

An integrated eco-evolutionary framework to predict population-level responses of climate-sensitive pathogens

It is critical to gain insight into how climate change impacts evolutionary responses within climate-sensitive pathogen populations, such as increased resilience, opportunistic responses and the emergence of dominant variants from highly variable genomic backgrounds and subsequent global dispersal. This review proposes a framework to support such analysis, by combining genomic evolutionary analysis with climate time-series data in a novel spatiotemporal dataframe for use within machine learning applications, to understand past and future evolutionary pathogen responses to climate change. Recommendations are presented to increase the feasibility of interdisciplinary applications, including the importance of robust spatiotemporal metadata accompanying genome submission to databases. Such workflows will inform accessible public health tools and early-warning systems, to aid decision-making and mitigate future human health threats.

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.

Machine learning based efficient prediction of positive cases of waterborne diseases

BACKGROUND

Water quality has been compromised and endangered by different contaminants due to Pakistan's rapid population development, which has resulted in a dramatic rise in waterborne infections and afflicted many regions of Pakistan. Because of this, modeling and predicting waterborne diseases has become a hot topic for researchers and is very important for controlling waterborne disease pollution.

METHODS

In our study, first, we collected typhoid and malaria patient data for the years 2017-2020 from Ayub Medical Hospital. The collected data set has seven important input features. In the current study, different ML models were first trained and tested on the current study dataset using the tenfold cross-validation method. Second, we investigated the importance of input features in waterborne disease-positive case detection. The experiment results showed that Random Forest correctly predicted malaria-positive cases 60% of the time and typhoid-positive cases 77% of the time, which is better than other machine-learning models. In this research, we have also investigated the input features that are more important in the prediction and will help analyze positive cases of waterborne disease. The random forest feature selection technique has been used, and experimental results have shown that age, history, and test results play an important role in predicting waterborne disease-positive cases. In the end, we concluded that this interesting study could help health departments in different areas reduce the number of people who get sick from the water.

Mathematical Models for Cholera Dynamics-A Review

Cholera remains a significant public health burden in many countries and regions of the world, highlighting the need for a deeper understanding of the mechanisms associated with its transmission, spread, and control. Mathematical modeling offers a valuable research tool to investigate cholera dynamics and explore effective intervention strategies. In this article, we provide a review of the current state in the modeling studies of cholera. Starting from an introduction of basic cholera transmission models and their applications, we survey model extensions in several directions that include spatial and temporal heterogeneities, effects of disease control, impacts of human behavior, and multi-scale infection dynamics. We discuss some challenges and opportunities for future modeling efforts on cholera dynamics, and emphasize the importance of collaborations between different modeling groups and different disciplines in advancing this research area.

A High-Resolution Earth Observations and Machine Learning-Based Approach to Forecast Waterborne Disease Risk in Post-Disaster Settings

Responding to infrastructural damage in the aftermath of natural disasters at a national, regional, and local level poses a significant challenge. Damage to road networks, clean water supply, and sanitation infrastructures, as well as social amenities like schools and hospitals, exacerbates the circumstances. As safe water sources are destroyed or mixed with contaminated water during a disaster, the risk of a waterborne disease outbreak is elevated in those disaster-affected locations. A country such as Haiti, where a large quantity of the population is deprived of safe water and basic sanitation facilities, would suffer more in post-disaster scenarios. Early warning of waterborne diseases like cholera would be of great help for humanitarian aid, and the management of disease outbreak perspectives. The challenging task in disease forecasting is to identify the suitable variables that would better predict a potential outbreak. In this study, we developed five (5) models including a machine learning approach, to identify and determine the impact of the environmental and social variables that play a significant role in post-disaster cholera outbreaks. We implemented the model setup with cholera outbreak data in Haiti after the landfall of Hurricane Matthew in October 2016. Our results demonstrate that adding high-resolution data in combination with appropriate social and environmental variables is helpful for better cholera forecasting in a post-disaster scenario. In addition, using a machine learning approach in combination with existing statistical or mechanistic models provides important insights into the selection of variables and identification of cholera risk hotspots, which can address the shortcomings of existing approaches.

Computer vision for detecting field-evolved lepidopteran resistance to Bt maize

BACKGROUND

Resistance evolution of lepidopteran pests to Bacillus thuringiensis (Bt) toxins produced in maize and cotton is a significant issue worldwide. Effective toxin stewardship requires reliable detection of field-evolved resistance to enable the implementation of mitigation strategies. Currently, visual estimates of maize injury are used to document changing susceptibility. In this study, we evaluated an existing maize injury monitoring protocol used to estimate Bt resistance levels in Helicoverpa zea (Lepidoptera: Noctuidae).

RESULTS

We detected high interobserver variability across multiple injury metrics, suggesting that the precision and accuracy of maize injury detection could be improved. To do this, we developed a computer vision-based algorithm to measure H. zea injury. Algorithm estimates were more accurate and precise than a sample of human observers. Moreover, observer estimates tended to overpredict H. zea injury, which may increase the false-positive rate, leading to prophylactic insecticide application and unnecessary regulatory action.

CONCLUSIONS

Automated detection and tracking of lepidopteran resistance evolution to Bt toxins are critical for genetically engineered crop stewardship to prevent the use of additional insecticides to combat resistant pests. Advantages of this computerized screening are: (i) standardized Bt injury metrics in space and time, (ii) preservation of digital data for cross-referencing when thresholds are reached, and (iii) the ability to increase sample sizes significantly. This technological solution represents a significant step toward improving confidence in resistance monitoring efforts among researchers, regulators and the agricultural biotechnology industry.

Climate Precursors of Satellite Water Marker Index for Spring Cholera Outbreak in Northern Bay of Bengal Coastal Regions

Cholera is a water-borne infectious disease that affects 1.3 to 4 million people, with 21,000 to 143,000 reported fatalities each year worldwide. Outbreaks are devastating to affected communities and their prospects for development. The key to support preparedness and public health response is the ability to forecast cholera outbreaks with sufficient lead time. How Vibrio cholerae survives in the environment outside a human host is an important route of disease transmission. Thus, identifying the environmental and climate drivers of these pathogens is highly desirable. Here, we elucidate for the first time a mechanistic link between climate variability and cholera (Satellite Water Marker; SWM) index in the Bengal Delta, which allows us to predict cholera outbreaks up to two seasons earlier. High values of the SWM index in fall were associated with above-normal summer monsoon rainfalls over northern India. In turn, these correlated with the La Niña climate pattern that was traced back to the summer monsoon and previous spring seasons. We present a new multi-linear regression model that can explain 50% of the SWM variability over the Bengal Delta based on the relationship with climatic indices of the El Niño Southern Oscillation, Indian Ocean Dipole, and summer monsoon rainfall during the decades 1997–2016. Interestingly, we further found that these relationships were non-stationary over the multi-decadal period 1948–2018. These results bear novel implications for developing outbreak-risk forecasts, demonstrating a crucial need to account for multi-decadal variations in climate interactions and underscoring to better understand how the south Asian summer monsoon responds to climate variability.