Early diagnosis and prognosis of impending critical syndromes are crucial for optimising therapeutic strategies in intensive care. The extensive data in the intensive care unit (ICU) can use machine learning models (ML) as the basis for patient-centred digital twins to manage the complexity and instability of critically ill patients’ health conditions.

Building machine learning models in the ICU is supported by the Diagnostic Expert Advisor (DEA), a research platform that transforms heterogeneous ICU data into ML models.

In the context of acute respiratory distress syndrome (ARDS), the feasibility of a prediction model for the syndrome developed within the DEA platform was demonstrated.

University Hospital Aachen (RWTH) is bringing its experience in embedding DEA into the open EDITH platform, so that it could open the way for the prediction of other critical conditions in ICU patients, such as septic shock, acute kidney injury.

For more information see here