Identifying Distinct, Effective Treatments for Acute Hypotension with SODA-RL: Safely Optimized Diverse Accurate Reinforcement Learning
This addresses the challenge of treatment selection in life-threatening hypotension for critical care providers, offering multiple plausible options rather than a single best strategy, though it is incremental as it builds on existing reinforcement learning approaches.
The paper tackled the problem of identifying distinct and effective treatment options for acute hypotension in critical care, where it's often unclear which interventions to use. The result was SODA-RL, a method that learned policies performing comparably to observed physician behaviors on a cohort of 10,142 ICU stays, providing different plausible alternatives.
Hypotension in critical care settings is a life-threatening emergency that must be recognized and treated early. While fluid bolus therapy and vasopressors are common treatments, it is often unclear which interventions to give, in what amounts, and for how long. Observational data in the form of electronic health records can provide a source for helping inform these choices from past events, but often it is not possible to identify a single best strategy from observational data alone. In such situations, we argue it is important to expose the collection of plausible options to a provider. To this end, we develop SODA-RL: Safely Optimized, Diverse, and Accurate Reinforcement Learning, to identify distinct treatment options that are supported in the data. We demonstrate SODA-RL on a cohort of 10,142 ICU stays where hypotension presented. Our learned policies perform comparably to the observed physician behaviors, while providing different, plausible alternatives for treatment decisions.