자료유형  

 학위논문

Control Number  

0017162304

International Standard Book Number  

9798382833071

Dewey Decimal Classification Number  

658

Main Entry-Personal Name  

Zhang, Suyanpeng.

Publication, Distribution, etc. (Imprint  

[S.l.] : University of Southern California., 2024

Publication, Distribution, etc. (Imprint  

Ann Arbor : ProQuest Dissertations & Theses, 2024

Physical Description  

170 p.

General Note  

Source: Dissertations Abstracts International, Volume: 85-12, Section: B.

General Note  

Advisor: Suen, Sze-Chuan.

Dissertation Note  

Thesis (Ph.D.)--University of Southern California, 2024.

Summary, Etc.  

요약Repeated decision-making problems in the context of uncertainty naturally arise in healthcare settings. Markov decision processes (MDPs) have proven useful in many healthcare contexts, integrating disease progression, decision-making, costs, and benefits into an optimization framework. However, implementing MDPs in healthcare settings is nontrivial due to challenges including incorporating unique characteristics of certain diseases, determining the optimal frequency of decision-making, and dealing with the infinite number of possible states.In this dissertation, we focus on specific healthcare problems and identify key structural properties to address healthcare questions. We present a finite horizon MDP framework for patients with acute liver failure in need of a transplant, determining the optimal timing for accepting a suboptimal organ to maximize one-year survival probability. Additionally, we study the value provided by having additional decision-making opportunities in each epoch. We provide structural properties of the optimal policies and quantify the difference in optimal values between MDP problems of different decision-making frequencies. We analyze a numerical example using liver transplantation in high-risk patients and treatment initiation for chronic kidney disease patients to illustrate our findings. Finally, in the fourth chapter, to address the curse of dimensionality, we propose a novel greedy algorithm for non-uniform discretization in a population-level MDP for infectious disease control.The dissertation contributes to the field of healthcare applications by providing practical MDP frame-works and efficient algorithms to tackle complex decision-making problems. The theoretical results and empirical analyses offer valuable guidance for healthcare decision-makers in diverse scenarios.

Subject Added Entry-Topical Term  

Industrial engineering.

Subject Added Entry-Topical Term  

Medicine.

Index Term-Uncontrolled  

Dynamic programming

Index Term-Uncontrolled  

Infectious disease control

Index Term-Uncontrolled  

Liver transplant

Index Term-Uncontrolled  

Healthcare settings

Index Term-Uncontrolled  

Decision-making

Added Entry-Corporate Name  

University of Southern California Industrial and Systems Engineering

Host Item Entry  

Dissertations Abstracts International. 85-12B.

Electronic Location and Access  

로그인을 한후 보실 수 있는 자료입니다.

Control Number  

joongbu:654866