Learning Health Systems (LHS) aim to operationalize continuous improvement by learning from prior practice to enhance future care. However, deep reflection on medical decision-making is currently a resource-intensive, manual process. This project proposes a novel framework: using Reinforcement Learning (RL) to build the analytical infrastructure required for systematic, expedited learning. This infrastructure provides the foundation to drive and measure continuous improvements, effectively closing the LHS loop between institutional data and improved patient outcomes.
