Antimicrobial resistance is one of the greatest challenges facing modern medicine. An increasing number of bacteria no longer respond to commonly used antibiotics. Already today, millions of deaths worldwide are associated with such resistance, and this number could continue to rise in the coming decades.

To treat patients effectively, the appropriate antibiotic must be administered as quickly as possible. However, current laboratory tests often take one to two days. For critically ill patients, this is frequently too slow.

Although new technologies and methods can accelerate diagnostics, they have not yet been fully integrated into routine clinical practice everywhere. This is where artificial intelligence and machine learning offer new opportunities: they can analyze large volumes of clinical and microbiological data to predict which antibiotic is likely to be effective. Instead of merely distinguishing between “effective” and “ineffective,” it is particularly useful to predict the exact required drug concentration (minimum inhibitory concentration). This enables more precise and better comparable treatment decisions, independent of changing clinical breakpoints.

In this project, the IMI is responsible for developing a data protection and IT security concept and for establishing a pseudonymized research database. The project is being carried out in cooperation with the Institute of Medical Microbiology and the Institute of Biostatistics and Clinical Research at the University of Münster.

Contact: Antje Westendorf, M.Sc.