This technology is a machine-learning-based system that estimates intravascular volume status in real time for patients receiving extracorporeal membrane oxygenation (ECMO).
Fluid overload is a major contributor to morbidity and mortality in critically ill children, particularly those in the pediatric intensive care unit requiring extracorporeal membrane oxygenation (ECMO) support. Despite the importance of maintaining optimal intravascular volume, there is no reliable, validated, objective measure to guide fluid therapy, and clinicians must rely on indirect parameters, circuit pressures, and subjective clinical assessment. Drainage pressure is commonly used in decision-making, but it has not been rigorously validated as a true indicator of intravascular volume. This gap increases the risk of inappropriate fluid administration, which can compromise ECMO performance and worsen patient outcomes.
The technology is a machine-learning system that predicts intravascular volume status in patients on extracorporeal membrane oxygenation (ECMO). It integrates real-time clinical data and ECMO circuit pressure measurements to provide patient-specific guidance for fluid management. The system is based on mathematical models developed from in vitro experiments, and clinical use will help further refine and improve these models. These models have the potential to be integrated into existing ICU data management platforms to support clinical decision-making.
IR CU25361
Licensing Contact: Joan Martinez