On the Challenges of using Reinforcement Learning in Precision Drug Dosing: Delay and Prolongedness of Action Effects

TL;DR

This paper addresses the challenges of applying reinforcement learning to drug dosing by tackling delayed and prolonged effects that violate the Markov assumption, proposing a method to convert such problems into standard MDPs for effective solution.

Contribution

The paper introduces PAE-POMDP, a subclass of POMDPs for prolonged action effects, and proposes a conversion method to restore the Markov property, enabling RL application in drug dosing.

Findings

The proposed method significantly outperforms vanilla RL baselines.

It is competitive with recurrent policies but more efficient.

The approach is suitable for real-time drug dosing systems.

Abstract

Drug dosing is an important application of AI, which can be formulated as a Reinforcement Learning (RL) problem. In this paper, we identify two major challenges of using RL for drug dosing: delayed and prolonged effects of administering medications, which break the Markov assumption of the RL framework. We focus on prolongedness and define PAE-POMDP (Prolonged Action Effect-Partially Observable Markov Decision Process), a subclass of POMDPs in which the Markov assumption does not hold specifically due to prolonged effects of actions. Motivated by the pharmacology literature, we propose a simple and effective approach to converting drug dosing PAE-POMDPs into MDPs, enabling the use of the existing RL algorithms to solve such problems. We validate the proposed approach on a toy task, and a challenging glucose control task, for which we devise a clinically-inspired reward function. Our results demonstrate that: (1) the proposed method to restore the Markov assumption leads to significant improvements over a vanilla baseline; (2) the approach is competitive with recurrent policies which may inherently capture the prolonged effect of actions; (3) it is remarkably more time and memory efficient than the recurrent baseline and hence more suitable for real-time dosing control systems; and (4) it exhibits favorable qualitative behavior in our policy analysis.