Global climate change is redefining our understanding of how diseases spread. In Sri Lanka, vector-borne diseases such as dengue fever historically surged during the monsoon seasons when temperatures were high enough for mosquito eggs to hatch. Unfortunately, due to rising temperatures and more erratic rainfall patterns, mosquito eggs can now hatch year-round making outbreaks increasingly unpredictable, leading to an alarming rise in hospitalizations and deaths. More data is needed to adapt our response to these diseases in an increasingly warmer world. In the contemporary landscape, a wealth of disease information is available, yet accessibility remains limited due to unstructured data formats such as PDFs. Therefore, converting unstructured disease reports into structured formats is necessary for effectively leveraging data. This paper introduces a comprehensive framework for collecting unstructured disease reports and transforming them into analyzable formats. By creating separate models tailored to each data format, we can ensure accuracy compared to general models. These straightforward models enhance accessibility and empower other researchers to use our tools. The returned structured data can then be harnessed for analysis, statistical purposes, and informing evidence-based public health interventions, thus facilitating more informed decision-making in healthcare. We deploy this framework to produce geospatial data for Sri Lanka and Brazil for many different conditions and align these data with satellite environmental data, providing for the first time a structured, aligned powerful dataset for disease modeling.

The global expansion of the dengue belt, driven by climate change and increased urbanization, has led to a significant rise in dengue cases worldwide (1). Early warning systems (EWS) coupled with prompt public health response mechanisms are crucial in mitigating dengue-related morbidity and mortality globally. In Sri Lanka, dengue transmission occurs year-round with two peaks correlating to the southwest monsoon from May to September and the northeast monsoon from October to January (2). The presence of multiple dengue virus serotypes (DENV1–4) complicates epidemiological patterns, as sequential infections with different serotypes can increase the risk of severe disease manifestations detected by surveillance systems (3). Understanding and integrating these virological dynamics, vector dynamics, and real-time surveillance data are essential for developing effective EWS and targeted public health interventions. We propose the use of Graph Neural Networks (GNNs) as an EWS. Using Earth observational data from NASA’s global satellites and dengue incidence data from Sri Lanka’s Ministry of Health, we developed traditional and graph-based EWS to forecast dengue cases across Sri Lanka’s 25 districts between 2013 and 2022. We demonstrate empirically that GNNs incorporating spatiotemporal relations significantly outperform traditional EWS models such as Autoregressive Integrated Moving Average (ARIMA), Random Forest, and Long Short-Term Memory (LSTM). Our source code is available on GitHub.

Electricity has become an indispensable part of our society. As natural disasters strike communities, it is crucial to prepare the power grid for the incoming storm. This exploratory paper delves into the methods used to predict power outages in the United States’ southeastern counties after hurricanes. The prediction is made using a range of machine learning techniques. Four methods, namely Multivariable Linear Regression, Random Forest, Extreme Gradient Boosting (XGBoost), and K-Nearest Neighbors Regression (KNN), are employed to predict the percentage of electric customers in a county that will face a power outage the next day. The accuracy of these models is assessed using regression metrics such as Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE). To assess a county’s vulnerability to power outages, meteorological, electrical, and socioeconomic data were taken into account. Rolling average data was also included for day-to-day features. A correlation matrix was used to select variables relevant to this analysis. The hyperparameters of each model were chosen based on the parameters that resulted in the lowest MSE in 5-fold cross-validation. After comparing different models, it was observed that the Random Forest Regression method had the lowest MSE of 5.93E-05, indicating sufficiency. These model implementations helped to better understand the nature of the problem and plan future research work toward predicting weather-based power outages, particularly those caused by hurricanes as they are related to precipitation (rainfall) which we included in this study. This analysis suggests that creating a model specifically for predicting power outages is necessary to understand what are the most influential weather-related variables that can be used in data-driven analysis.