Project R-16870

Bayesian Modeling of Epidemiological Delays (BAYESMED) (Research)

The recent COVID-19 pandemic plunged the world into a profound crisis and took a heavy toll on global health and socio-economic conditions. The aftermath of such stress testing periods reveals a fertile soil to develop new methodologies aimed at extracting meaningful information from infectious disease data. Statistical modeling is of paramount importance to analyze epidemic data collected by public health systems since inferential methods permit to estimate epidemiological factors that are used to cast light on the disease dynamics of an invading pathogen. Epidemiological delays (times between two well defined events related to a disease) play a key role to understand disease transmission. Information on delay quantities can be leveraged to mitigate the public health impact caused by epidemic and pandemic-prone pathogenic organisms. Three shortcomings are identified in the state-of-the-art research landscape on epidemiological delays; namely, parametric assumptions (PARA), computational aspects for inference (COMP) and support for reproducible modeling (RMOD). Innovative ideas anchored around semi-parametric models, approximate Bayesian inference and functional programming paradigms are proposed to address these limitations. The project will unlock the full potential of Bayesian methods through a reliable, fast and reproducible suite of tools for modelling epidemiological delays and bridge important scientific gaps by using ground-breaking concepts and strategies.

01 October 2026 - 30 September 2029