Improvement of the software for modeling the dynamics of epidemics and developing a user-friendly interface

Area of Science:

• Epidemiology
• Mathematical Modeling
• Public Health

Background:

• Accurate epidemic forecasting is crucial for managing public health and economic challenges posed by pandemics like COVID-19.
• Complex models require extensive data and parameter estimation, often limited by available observations.
• The basic SIR (Susceptible-Infectious-Recovered) model, despite its simplicity, has shown success in modeling early COVID-19 waves.

Purpose of the Study:

• To adapt and apply the SIR model for accurate epidemic forecasting, even with limited data.
• To develop a generalized SIR model capable of simulating various epidemic waves and accounting for data limitations.
• To create a user-friendly interface for continuous epidemic monitoring, prediction, and comparative analysis.

Main Methods:

• Utilized the SIR model with an original parameter identification algorithm for initial COVID-19 wave analysis.
• Proposed a generalized SIR model and associated algorithms to handle diverse epidemic dynamics and incomplete data.
• Developed a user-friendly interface for real-time SIR simulations and comparative analysis of epidemic data.

Main Results:

• The SIR model demonstrated high accuracy in predicting the duration and case numbers of early COVID-19 waves in several countries.
• Simulations of the COVID-19 wave in Japan (Summer 2022) showed good agreement between predicted and observed case numbers, particularly with recent data.
• Comparative analysis revealed that high vaccination levels did not prevent a significant 2022 wave in Japan, though the death-to-case ratio was lower than in 2020.

Conclusions:

• The SIR model, particularly the generalized version with a robust parameter identification algorithm, is effective for epidemic forecasting.
• A user-friendly interface is essential for frequent monitoring and timely predictions, enabling better public health responses.
• Epidemic dynamics are influenced by multiple factors including vaccination, quarantine, and social behavior, necessitating continuous analysis and comparison.