Imitation learning for clinical decision support in pediatric ECMO

ArXi:2605.16175v1 Announce Type: new Pediatric critical care is a dynamic, high-stakes process involving constant monitoring and adjustments in life-saving treatments. Modeling these interventions is crucial for effective decision. To address the challenges of high complexity and data scarcity in pediatric Extracorporeal Membrane Oxygenation (ECMO), we frame clinical decision-making as learning to act from trajectories, i.e., imitation learning that learns action models from observational data, with a key feature that actions are not directly observed.