Medical decision support for treating infectious diseases
Ladda ner
Publicerad
Författare
Typ
Examensarbete för masterexamen
Modellbyggare
Tidskriftstitel
ISSN
Volymtitel
Utgivare
Sammanfattning
With an ever increasing digitalization of health care and records thereof, possibilities
of applying data driven methods to analyze and predict healthcare outcomes
are becoming more numerous. Markov decision processes (MDPs) are an
interesting class of models to explore in this context. Originally born out of the
operations research field, they allow us to model sequential decision making under
uncertainty. Furthermore they may be extended to partially observable MDPs
(POMDPs), in order to model further uncertainty, such as decision making with
incomplete information which is often the case in healthcare. In this project, a
literature review has been made in order to analyze and discuss mainly POMDPs
from the perspective of treating sepsis and COVID-19.
Beskrivning
Ämne/nyckelord
Partially observable Markov decision processes, Bayesian reinforcement learning