JMIR Publications

ABSTRACT

Background:

Reinforcement Learning (RL)-based decision support systems have been implemented to facilitate the delivery of personalised care. This paper details a systematic review of RL applications in the critical care setting.

Objective:

This review aimed to survey the literature with regards to RL applications for clinical decision support in critical care, and to provide insight into the challenges of applying various RL models.

Methods:

We performed a systematic search of the following databases: PubMed, Google Scholar, IEEE, ScienceDirect, Web of Science, MEDLINE and Embase. Studies published over the past 10 years (2010-2019) that have applied RL for critical care were included.

Results:

We included 21 papers. We found that RL has been used to optimise choice of medications, drug dosing and timing of interventions, and to target personalised laboratory values. We further compared and contrasted the design of the RL models and the evaluation metrics in each application.

Conclusions:

RL has great potential for enhancing decision-making in critical care. Challenges regarding RL system design, evaluation metrics and model choice exist. More importantly, further work is required to validate RL in authentic clinical environments.