Digital Twin Model for Venovenous Extracorporeal Membrane Oxygenation Based on Pulse Physiology Engine

Venovenous extracorporeal membrane oxygenation (VV-ECMO) is a form of mechanical respiratory support therapy for patients with acute lung injury or end-stage lung disease. Clinical demand for VV-ECMO has increased substantially, yet consensus on optimal patient management strategies remains limited. ECMO physiology is highly complex and difficult to predict, given the three-way interactions between the patient, the mechanical ventilator, and the extracorporeal circuit. A physiological simulation platform would be a valuable tool for clinical decision support during ECMO, but currently available models have limited capability to simulate high-resolution transient responses. To fill this gap, we present a digital twin model of ECMO based on the Pulse Physiology Engine platform. The model was successfully validated using ten VV-ECMO patient datasets from a prior clinical study. Simulated transient venous oxygenation profiles closely matched measurements from an independent porcine VV-ECMO study. We present a representative simulation scenario that illustrates how VV-ECMO support enables titration and optimization of lung-protective ventilator settings. This digital twin model of VV-ECMO, based on the Pulse Physiology Engine, provides a foundation for future development of clinical decision-support tools and automated extracorporeal life-support technologies.