Optimización automática de parámetros en simulaciones epidemiológicas de alta precisión

The health crisis that the world is experiencing due to the COVID-19 epidemic has had a huge impact on public
health, the economy, educational systems and many other fields. It is predicted that epidemics like this will
become the norm in the coming years due to the prevailing economic models that push the border with nature
more vigorously. The scientific community has sought to understand the dynamics of the current pandemic
from multiple fronts. In particular, epidemiological models have emerged as an indispensable tool in the
creation of public policies to face the changing circumstances of the pandemic in different regions of the world.
The simulations of epidemiological models are vital because they allow anticipating potentially catastrophic
consequences for health systems. Should a new quarantine be imposed with the latest rise in cases? When
could the educational system return in person? What would be the most effective vaccination strategy? These
questions are answered by simulations of epidemiological models. Several of these models have been tried in
Costa Rica, each with its advantages and disadvantages. An international collaboration between the
Tecnológico de Costa Rica, the Centro Nacional de Alta Tecnología (CeNAT), the Instituto Costarricense de
Investigación y Enseñanza en Nutrición y Salud (Inciensa) and the Federal Institute of Paraná in Brazil has
developed a highly accurate epidemiological model for Costa Rica, based on the Corona++ computational
code. These simulations are efficient to execute and allow the study of phenomena at very small scales, such
as contagions at the district level. This degree of precision allows us to better understand the dynamics of the
epidemic and how to establish control strategies. However, before running the simulations it is necessary to
adjust multiple parameters of the model. This process is manual, slow and tedious. Worse still, changes in the
premises of the model require a readjustment of the parameters. With this research project, we intend to
automatically optimize the parameter selection of the epidemiological model simulations in Corona++ to allow
the code to be, not only efficient as a whole, but also easily scalable to other environments and flexible to
study different variables of interest. The computational base of this project may be extended in the future to
other types of epidemics, new and already known. New coronaviruses could appear in the future and generate
epidemics of great proportions. Known epidemics periodically impact the country, such as seasonal influenza,
dengue, and Zika. Having a computational base for epidemics is crucial to guarantee a better response to
these epidemics and improve the quality of life of Costa Ricans.

Construir un optimizador automático de parámetros para simulaciones epidemiológicas de
alta precisión

1) Teoría y Metodologías en Computación
2) Aplicación de la computación en distintos dominios científicos, tecnológicos, organizacionales y sociales