Application of Deep Recurrent Q Network with Dueling Architecture for Optimal Sepsis Treatment Policy

• 발행기관: 한국스마트미디어학회
• 저자명: Thanh-Cong Do 양형정(Hyung Jeong Yang) Ngoc-Huynh Ho
• 간행물 정보: 『스마트미디어저널』Vol10, No.2, 48~54쪽, 전체 7쪽
• 주제분류: 공학 > 컴퓨터학
• 파일형태: PDF
• 발행일자: 2021.06.30

영문 초록

Sepsis is one of the leading causes of mortality globally, and it costs billions of dollars annually. However, treating septic patients is currently highly challenging, and more research is needed into a general treatment method for sepsis. Therefore, in this work, we propose a reinforcement learning method for learning the optimal treatment strategies for septic patients. We model the patient physiological time series data as the input for a deep recurrent Q-network that learns reliable treatment policies. We evaluate our model using an off-policy evaluation method, and the experimental results indicate that it outperforms the physicians’ policy, reducing patient mortality up to 3.04%. Thus, our model can be used as a tool to reduce patient mortality by supporting clinicians in making dynamic decisions.

목차

I. INTRODUCTION

II. BACKGROUND AND RELATED WORK

III. SETTING UP ENVIRONMENT FOR REINFORCEMENT LEARNING

IV. PROPOSED METHOD

V. EXPERIMENTAL RESULTS

키워드

Sepsis Reinforcement Learning Deep Learning Deep Recurrent Q Network