Individualized antibiotic treatments by integrating time-lapse microscopy, image analysis, machine-learning and PKPD modelling techniques

Antibiotic resistance is a major threat to human health, and a leading cause of death world-wide.

Compared to many other therapeutic areas, the use of personalized medicine in infectious diseases is lagging behind, and most antibiotic treatments are still delivered according to a “one size fits all” approach.

To address this, we will in this multidisciplinary research program develop a novel combinatorial framework that will provide unique insights into the dynamics of the bug-drug interaction with the aim to support individualized antibiotic treatments based on the bacterial suseptibility profile of the infecting pathogen.

Image analysis and machine learning methods will be used to efficiently quantify bacterial densities and antibiotic-induced morphological changes in time-lapse microscopy data for antibiotics used alone and in combination.

Mathematical PKPD modelling will be used to quantitatively understand the dynamic link between drug exposure, bacterial growth and killing, and the emergence of resistance.

The models will be applied in simulations to identify key bacteria and patient specific factors of importance for personalized treatments in critically ill patients with multidrug-resistant bacteria.

This unique combination of methods will serve as a tool for tailoring antibiotic treatments to the individual patient, with the aim to improve treatment outcome, slow down resistance development and extend the effective life-span of both existing and future antibiotics.