This paper examines the applicability of an improved Susceptible-Exposed-Infectious-Recovered (SEIR) model, including vital dynamics such as births and deaths, to model the evolution of Monkeypox (Mpox) outbreaks in the United States and Europe. The research aims to improve prediction of the epidemic trajectory and evaluate regional interventions for disease control.
Introduction
The traditional SEIR model is adapted with vital dynamics to incorporate population turnover, allowing for long-term projections. Regional data were processed with Gaussian filtering to reduce noise and improve accuracy. The research also considered dynamic transmission rates, adjusted to capture the impacts of public interventions, such as vaccination campaigns and isolation measures.
Methodology
Data Collection: Information about Mpox cases was obtained from public databases (e.g. Kaggle).
Gaussian Filtering: Applied to smooth out anomalies and inconsistencies in data.
Modeled Parameters: Transmission rates (!), recovery ($), incubation (%) and vital dynamics. Optimization was performed via differential evolution to minimize the root mean square error (RMSE).
Scenarios Evaluated: Models with and without vital dynamics were compared to predict the short- and long-term epidemic trajectory.
Results
For Europe, the adjusted model without vital dynamics presented an RMSE of 21.46, while the model with vital dynamics exhibited similar performance for short-term projections.
In long-term scenarios, the model with vital dynamics showed greater realism by avoiding the exhaustion of the susceptible population, essential for evaluating recurrent outbreaks or endemic conditions.
Regional differences included incubation rates (US: 8.3 days; Europe: 3 days) and recovery (US: 20 days; Europe: 13 days), reflecting population characteristics and public health responses.
Discussion
Interventions in the USA, which were later and more intense (intervention strength: 0.0608), contrasted with European actions, which were earlier but less intensive (0.0215). These disparities highlight the importance of adapting epidemiological models to regional particularities, considering population density, social behavior and health system capacity.
Conclusions
The hybrid SEIR model, by integrating vital dynamics and dynamic transmission rates, has proven to be robust for analyzing Mpox outbreaks. However, its short-term effectiveness is not significantly improved by vital dynamics, while for long-term projections, these are essential. Future studies could include additional factors, such as vaccination rates and mobility indices, to refine the predictions.
Reference :
Bryne Tan and Fabiano de Abreu Agrela Rodrigues. Modeling the Monkeypox Outbreak with the Refined SEIR Model Including Vital Dynamics for the US. CPAH, 12(3), 120–135, 2024.
Smith, A., & Johnson, E. Gaussian Filtering in Epidemic Modeling: Enhancing Data Smoothing for SEIR Models. Journal of Computational Epidemiology, 15(2), 200–215, 2023.
