RL4CAD

RL4CAD: Personalized Decision Making for Coronary Artery Disease Treatment using Offline Reinforcement Learning

This repository contains the codes and resources related to the following paper:

@article{ghasemi2024RL4CAD,
title={Personalized Decision Making for Coronary Artery Disease Treatment using Offline Reinforcement Learning},
author={Ghasemi, Peyman and White, James A and Lee, Joon},
year={2024},
journal={Preprint}
}

Abstract

Choosing optimal revascularization strategies for patients with obstructive coronary artery disease (CAD) remains a clinical challenge. While randomized controlled trials offer population-level insights, gaps remain in the literature regarding personalized decision-making for individual patients. We applied off-policy reinforcement learning (RL) in an offline setting to estimate optimal treatment policies for obstructive CAD and evaluated these policies using weighted importance sampling. Our findings indicate that RL policies generally outperformed physician-based strategies, with the best RL policy achieving about a 38% improvement in expected rewards based on a composite major cardiovascular events outcome. Additionally, we introduced methods to ensure RL CAD treatment policies remain compatible with existing clinical practices and presented an interpretable RL policy with a limited number of states. We anticipate our models based on these methods, RL4CAD, will facilitate optimal CAD treatment decision-making.

Model Weights

https://drive.google.com/drive/folders/1SuDCfoNeZWBifAFqBurDAbgfR7Xm2m5x?usp=sharing