讲座题目：PenalizedMixture Cure Rate and Promotion Time Cure Rate Models with Image Predictors

主讲人：俞章盛 上海交通大学

讲座时间：2020年6月24日（周三）下午3：00—4：00

讲座地点：腾讯会议ID： 115 786 197

主讲人简介：

俞博士，上海交通大学生物信息及生物统计系长聘教授，耶鲁大学生物统计系兼职教授，交大-耶鲁联合生物统计与数据科学中心副主任。美国密歇根大学统计学博士。 目前担任两个统计学SCI杂志Associate Editor， 及被Heart Rhythm(影响因子5.0)和pediatric pulmonology(2.7)聘为Statistical Editor 或Editorial board member。主要研究方向：临床试验设计，生存分析，纵向数据分析，非参数统计，生物信息学，高通量数据分析等。

讲座摘要：

Survival data including potentially cured subjects are common in clinical studies and mixture cure rate models are often used for analysis. The non-cured probabilities are often predicted by non-parametric, high-dimensional, or even unstructured (e.g. image) predictors, which is a challenging for traditional nonparametric methods such as spline and local kernel. For the mixture cure rate model and promotion time cure rate model, we developed the algorithm to incorporate the neural network to model the nonparametric or unstructured predictors’ effect in cure rate models. We estimate the parameters by Expectation-Maximization (EM) algorithm. Asymptotic convergence rates of our estimators are established. Simulation studies show good performance in both prediction and estimation. Finally, we analyze Primary Biliary Cirrhosis (PBC) data and Open Access Series of Imaging Studies (OASIS) data to illustrate the practical use of our methods. [Yujing Xie and Zhangsheng Yu*]