Methodology for assessing the impact of emergencies on the spread of infectious diseases

The spread of infectious diseases is significantly influenced by emergencies, particularly military conflicts, which disrupt healthcare systems and increase the risks of epidemics. The full-scale Russian invasion of Ukraine has exacerbated these challenges, causing environmental damage, mass displacement, and the breakdown of healthcare services, all of which contribute to the spread of infectious diseases. This study aims to develop a comprehensive methodology for assessing the impact of emergencies on the spread of infectious diseases, focusing on the full-scale invasion of Ukraine. The object of this study is to address epidemic threats posed by emergencies, particularly the increased spread of infectious diseases due to war-related disruptions. The subject of this study is methods and models of infectious disease transmission under conditions of emergencies, emphasizing the Russian full-scale invasion of Ukraine. The tasks of this study are to provide an analysis of the current state of research and develop a methodology for assessing the impact of emergencies on the spread of infectious diseases. The proposed methodology includes several key components. Comprehensive data from public health organizations includes infectious disease statistics, demographic shifts, healthcare disruptions, and environmental factors exacerbated by emergencies. Data preprocessing removes inconsistencies, standardization of formats, and normalization for population size differences. Machine learning models, including convolutional neural networks and recurrent neural networks, have been developed to simulate the spread of diseases based on demographic, environmental, and healthcare-related variables. Deep learning models analyze spatial and temporal patterns, whereas compartmental models such as SIR estimate changes in reproductive numbers (R₀ and Re). Additionally, models of excess mortality incorporate mixed effects to account for regional and time-based variations. The methodology incorporates real-time monitoring of epidemic threats using real-time data from multiple sources, enabling dynamic assessments of disease spread and facilitating predictive modeling. The models were trained on historical data and validated using cross-validation techniques to ensure robustness and reliability, with a specific focus on the pre- and post-invasion phases in Ukraine.Results: The study provides a comprehensive framework for collecting and processing data on infectious diseases and epidemic threats in emergencies. The proposed model introduces advanced machine learning and epidemiological models trained on pre- and post-invasion data to analyze disease transmission patterns and forecast future epidemic dynamics. Conclusion: The proposed methodology addresses current gaps in infectious disease during emergencies by integrating real-time data and machine learning techniques. This research improves decision-making in public health management and biosafety during crises, particularly in war-affected regions like Ukraine.