ICU sedation, delirium & recovery
Does Reinforcement Learning Improve Outcomes for Critically Ill Patients? A Systematic Review and Level-of-Readiness Assessment
Authors' conclusion
In this first systematic review on the application of RL in intensive care medicine we found no studies that demonstrated improved patient outcomes from RL-based technologies.
Study context
- Population
- Critically ill (ICU) patients18 journal + 18 conference papers
- Comparison
- Reinforcement learning (RL)-agent policies vs clinician policies
- Outcome
- Patient health-related outcomes of RL vs clinician's policy
Methodological detailsrisk of bias and applicability assessment, 1,033 articles screened; search to March 25, 2022
- Risk of bias and applicability assessment
- 1,033 articles screened; search to March 25, 2022
Pooled result
As reported in the abstract
- No studies demonstrated improved patient outcomes from RL-based technologies
- All articles reported RL algorithms to outperform clinical decision-making, but only in retrospective data (off-policy evaluation)
Further pooled results
Level of readiness
- Prototyping or modeling articles
- 30
- Validation articles
- 6
Risk of bias
- High in all articles, mainly due to evaluation procedure
Original publication
Does Reinforcement Learning Improve Outcomes for Critically Ill Patients? A Systematic Review and Level-of-Readiness Assessment. Crit Care Med. Published 8 November 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →