Statistical Model for Predicting Survival Rate among Acute Myeloid Leukemia Patients after Performing Allogeneic Stem Cell Transplantation operation

Abstract

Survival analysis is generally defined as a set of statistical procedures for analyzing data for which the outcome variable of interest is time until an event occurs. In a survival analysis, we usually refer to the dependent variable as survival time, as it gives the time that an individual has survived over some follow-up period. We also commonly refer to the event as a failure. Most survival analyses must take into consideration a key analytical problem called censoring. Censoring generally occurs when some subjects in the study has not experienced the event of interest at the end of the study or time of analysis. Cox regression is considered the standard method for analyzing censored data, but the assumptions required of these models which assume that the hazard function of any two levels of some covariates is proportional over the period of follow up period are easily violated. Several strategies have been developed to extend machine learning techniques to accommodate right censored data. In this study, we introduce also classification tree analysis as a flexible alternative for modeling censored data. Classification tree analysis is a decision-tree model that provides decision rules that maximize predictive accuracy.