ICU-Sepsis: A Benchmark MDP Built from Real Medical Data

TL;DR

ICU-Sepsis is a standardized, real-data-based MDP environment designed for benchmarking reinforcement learning algorithms in the complex task of sepsis management in ICU settings.

Contribution

The paper introduces ICU-Sepsis, a lightweight, tabular MDP environment derived from real patient data, for evaluating RL algorithms in sepsis care.

Findings

ICU-Sepsis is challenging for state-of-the-art RL algorithms.

It provides a standardized platform for benchmarking RL performance.

The environment models personalized ICU sepsis management based on real data.

Abstract

We present ICU-Sepsis, an environment that can be used in benchmarks for evaluating reinforcement learning (RL) algorithms. Sepsis management is a complex task that has been an important topic in applied RL research in recent years. Therefore, MDPs that model sepsis management can serve as part of a benchmark to evaluate RL algorithms on a challenging real-world problem. However, creating usable MDPs that simulate sepsis care in the ICU remains a challenge due to the complexities involved in acquiring and processing patient data. ICU-Sepsis is a lightweight environment that models personalized care of sepsis patients in the ICU. The environment is a tabular MDP that is widely compatible and is challenging even for state-of-the-art RL algorithms, making it a valuable tool for benchmarking their performance. However, we emphasize that while ICU-Sepsis provides a standardized environment for evaluating RL algorithms, it should not be used to draw conclusions that guide medical practice.