A0549
Title: Survival tree model for cure rate model with interval-censored data
Authors: Yang-Jin Kim - Sookmyung Women University (Korea, South) [presenting]
Abstract: A regression tree method is proposed for cure rate survival data when some patients are insusceptible and risk-free. Survival trees are fully nonparametric methods with graphical results and can easily handle high dimensional covariate space. A survival tree, composed of a cure group as well as heterogeneous subgroups, is constructed using a conditional inference tree for both right-censored data and interval-censored data. The performance of the proposed method is compared to another prediction model in terms of MSE and modified integrated Brier score (IBS). Eastern Cooperative Oncology Group (ECOG) dataset and Smoking cessation study are analyzed as real data examples.
