Knowledge Loops for the Containment of Drug-Resistance: Driving Change by Correlating Risk and Infections

The aim of the project is to use advanced statistical and machine learning techniques to interrogate the vast amount of existing data relating to antimicrobial resistance (AMR). It is of importance to identify key risk factors driving clinical cases of resistant infections in a One Health framework. The project relies on a network of researchers across Africa generating relevant data, which implies that the data will come in many different forms and wildly varying quality. This is seen as part of the key research question: instead of assessing the data from a binary perspective (i.e. "Yes, a correlation exists" or "No, there is no correlation"), we will ask "What is the most we can possibly get out of the data, and what is the most cost-effective way to improve prediction?" AMR surveillance will always be limited in resource constrained settings; our research will assess the potential of generated data and identify lowest hanging fruit within these constraints.