Optimizing Cancer Screening Using Partially Observable Markov Decision Processes

Abstract

Please scroll down for article-it is on subsequent pages With 12,500 members from nearly 90 countries, INFORMS is the largest international association of operations research (O.R.) and analytics professionals and students. INFORMS provides unique networking and learning opportunities for individual professionals, and organizations of all types and sizes, to better understand and use O.R. and analytics tools and methods to transform strategic visions and achieve better outcomes. For more information on INFORMS, its publications, membership, or meetings visit http://www.informs.org INFORMS 2011 c 2011 INFORMS | isbn 978-0-9843378-2-8 doi: http://dx. Abstract This tutorial describes the use of partially observable Markov decision processes (POMDPs) for optimizing cancer screening decisions. POMDP models can be used to address several controversial open research questions in cancer screening, such as when to start and stop screening and how often to screen. POMDP models provide a well-suited framework to optimize screening decisions because they allow the representation of the unobservable true health condition of a patient and screening tests that provide partial information about the true health condition. This tutorial uses a previously developed POMDP model for mammography screening to demonstrate the development and application of a POMDP model for cancer screening. In addition, challenges for applying POMDPs to model other cancer screening problems as well as possible future research directions are described.