Use Generalized Pareto Survival Models to Estimation Optimal Survival Time for Myocardial Infarction Patients

Abstract

AbstractThe survival analysis is one of the modern methods of analysis that is based on the fact that the dependent variable represents time until the event concerned in the study. There are many survival models that deal with the impact of explanatory factors on the likelihood of survival, including the models proposed by the world, David Cox, one of the most important and common models of survival, where it consists of two functions, one of which is a parametric function that does not depend on the survival time and the other a nonparametric function that depends on times of survival, which the Cox model is defined as a semi parametric model, The set of parametric models that depend on the time-to-event distribution parameters such as Exponential Model, Weibull Model, Log-logistic Model. Our research aims to adopt some of the Bayesian Optimal Criteria in achieving optimal design to estimate the optimal survival time for patients with myocardial infarction by constructing a parametric survival model based on the probability distribution of the survival times of myocardial infarction patients, which is among the most serious diseases that threaten human life and the main cause of death all over the world, as the duration of survival of patients with myocardial infarction varies with the factor or factors causing the injury, there are many factors that lead to the disease such as diabetes, high blood pressure, high cholesterol, psychological pressure and obesity. Therefore, the need to estimate the optimal survival time was expressed by constructing a model of the relationship between the factors leading to the disease and the patient survival time, and we found that the optimal rate of survival time is 18 days.