We use machine learning approaches in order to forecast the timing, intensity and duration of future epidemics. Mathematical models are developed and fitted to surveillance data in order to test hypotheses about mechanisms and drivers of transmission, and also to explore intervention strategies (in silico experiments) and look for optimal policies (optimal control theory). Our activities also involve theoretical and methodological developments.

We are generally interested in the seasonality of infectious diseases and its drivers (sociological, environmental, climatic), as well as in the spatial dynamics of infectious diseases, and its causes and consequences in terms of prevention and control. The research program is headed by Dr Marc Choisy and includes projects on influenza, dengue, measles (and other vaccine-preventable diseases), as well as COVID-19. The team has members of various backgrounds in biology, mathematics and computer science.

The team is also dedicated in running a national sero-surveillance system together with OUCRU Ha Noi, the Hospital of Tropical Diseases in Ho Chi Minh city and the National Institute of Hygiene and Epidemiology in Hanoi. The whole system is based on the exploitation of residual serum samples that are routinely collected in the hospitals. The analysis of such samples allows to assess – almost real-time – the population levels of protection against a panel of infectious diseases and identify gaps in vaccination. Such a system is other extremely useful to study any infectious diseases, even those that are not vaccine-preventable, and even those for which the serology is not correlation with protection. It can for example be used to assess the level of transmission of diseases that are fully (TB) or partially silently transmitted (e.g. dengue, SARS-CoV-2) and to understand better the determinants of population susceptibility to pathogens that are antigenically diverse such as dengue or influenza.