An individualized model-based "digital twin" for critical care decision support.

This project will support clinical decision-making in critical care by providing an individualized model of the cardiopulmonary system, a so-called digital twin. A digital twin can explain the complex underlying (patho)physiology and allows to safely test a proposed intervention on this model. The innovation of this project is to fully utilizing patient data with deep-learning algorithms combined with general models of cardiopulmonary physiology. The deep-learning algorithms will be used for system identification in order to transform general mathematical models of human physiology to the individual patient.