Improving Sepsis Treatment Strategies by Combining Deep and Kernel-Based Reinforcement Learning
This addresses the challenge of tailoring sepsis treatment for individual patients in critical care, representing an incremental improvement over existing methods.
The paper tackled the problem of personalizing sepsis treatment in the ICU by applying a mixture-of-experts framework that combines kernel-based and deep reinforcement learning, resulting in outperformance over physician, kernel-only, and DRL-only methods on a large retrospective cohort.
Sepsis is the leading cause of mortality in the ICU. It is challenging to manage because individual patients respond differently to treatment. Thus, tailoring treatment to the individual patient is essential for the best outcomes. In this paper, we take steps toward this goal by applying a mixture-of-experts framework to personalize sepsis treatment. The mixture model selectively alternates between neighbor-based (kernel) and deep reinforcement learning (DRL) experts depending on patient's current history. On a large retrospective cohort, this mixture-based approach outperforms physician, kernel only, and DRL-only experts.