Improving Sepsis Treatment Strategies by Combining Deep and Kernel-Based Reinforcement Learning

By Xuefeng Peng, David Wihl, Omer Gottesman, Matthieu Komorowski, Li-wei H. Lehman, Andrew Slavin Ross, Aldo A. Faisal, Finale Doshi-Velez. 2018

110 citations recorded by the source corpus.

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.

Subfields: Artificial Intelligence, Cardiology and Cardiovascular Medicine, Epidemiology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Cardiology and Cardiovascular Medicine

In Epidemiology

Citations in the corpus

Reinforcement Learning in Healthcare: A Survey (2019, methodology): “[178] applied the mixture-of-experts framework [176] in sepsis treatment by automatically switching between kernel learning and DRL depending on patients’ treatment history.”
POPCORN: Partially Observed Prediction COnstrained ReiNforcement Learning (2020, methodology): “POMDPs have been applied to heart disease management (Hauskrecht and Fraser, 2000), sepsis treatment in off-policy or simulated settings (Li et al., 2018; Oberst and Sontag, 2019; Peng et al., 2018), and HIV management (Parbhoo et al., 2017).”