MARKOV DECISION PROCESS MODEL FOR ASYMPTOMATIC INTRACRANIAL ANEURYSM TREATMENT SELECTION UNDER RISK AVERSION

摘要

We use a finite-horizon Markov Decision Process (MDP) model to find optimal treatment plans for patients diagnosed with asymptomatic intracranial aneurysm. We consider two distinct criteria: (a) maximizing the expected number of life years in good health, and (b) maximizing the expected utility for the number of years in good health. In (b) we assume exponential utility and consider different risk aversion factors. Incorporating risk aversion leads to "no treatment" recommendation for some situations. Use of alternate patient-selected criteria leads to recommendations that vary from a previously published Markov Process Monte Carlo simulation model [1] in several scenarios. The algorithm's computational performance makes it feasible for use in a software based decision support tool during a short office visit.