A swarm-optimizer-assisted simulation and prediction model for emerging infectious diseases based on SEIR

Area of Science:

• Epidemiology
• Computational Biology
• Public Health

Background:

• Current infectious disease prediction relies on mechanism-driven or statistical models.
• Integrating these approaches can yield more robust simulation and forecasting capabilities.

Purpose of the Study:

• To develop a hybrid model combining epidemic dynamics with data-driven parameter estimation for enhanced infectious disease prediction.
• To introduce a swarm-optimizer-assisted method for refining model parameters and forecasting disease spread.

Main Methods:

• A Susceptible-Exposed-Infected-Recovered (SEIR) model incorporating population migration was developed.
• A data-driven approach using public health and migration data was employed to determine model parameters via an objective function.
• A level-based learning swarm optimizer was utilized to optimize epidemic mechanism parameters.

Main Results:

• The proposed hybrid model demonstrated effectiveness in simulating and predicting emerging infectious diseases.
• Swarm optimization successfully refined model parameters, improving prediction accuracy.
• Experimental validation confirmed the efficacy of the integrated model and prediction method.

Conclusions:

• Combining mechanistic epidemic models with data-driven optimization offers a powerful approach for infectious disease forecasting.
• The developed swarm-optimizer-assisted method provides a robust tool for parameter estimation and prediction.
• This integrated strategy enhances the comprehensive understanding and management of emerging infectious diseases.