Development of Open-Source Real-Time DO2i Area Under the Curve (AUC) software for Goal-Directed Perfusion using Artificial Intelligence.

Background During cardiopulmonary bypass many parameters are observed to ensure metabolic demand is met. Among the most important of those are measures of oxygen supply and demand. Historically this would have included arterial blood-gas sampling, continuous venous oxygen saturation by optical cuvette, and possibly in-line blood or oxygenator gas measurement. Subsequently, additional metrics have been suggested to measure adequacy of perfusion. Yet some of the most promising, such as DO2i area under the curve (AUC) and the largest single AUC deficit episode, are often limited to retrospective statistical analysis and not accessible to clinicians. Methods We have developed non-commercial software that interfaces with commercially available perfusion monitors and can be run on most personal computers. The application displays real-time DO2i AUC and tracks deficits using both numeric and graphical visualizations. Results The software successfully imports DO2i results from an in-line blood monitor (Terumo OneView) and displays the DO2i AUC and largest single AUC deficit episode in real-time.