Stochastic Hybrid Algorithms to Estimate Stroke in Diabetic Patients

Abstract

Commonly, stroke is strongly related to those periods by which the patient has surpassed high values of glucose as well as when there is evidence of high cholesterol and blood pressure and others minor causes. While a permanent control inside the allowed ranges at the critic indicators is achieved, then the risk to acquire stroke turns out to be pretty low. In this paper, the theorem of Bayes and the Monte Carlo method are combined to construct a hybrid algorithm that yields the probability of stroke event as function of Bayes prediction and the Monte Carlo steps. The results are strongly depending on the life-style of patient so that the algorithm can be used to explore possible pessimistic scenarios that can be systematically optimized. From results of this paper, an event of stroke might emerge from a few Monte Carlo steps with a high Bayesian probability, fact that turns out to be highly dependent on conditions and history of patient more than random sequences.