学术报告
报告题目：Modeling and Prediction of Disease Processes Subject to Intermittent Observation
报 告 人： 武颖（ University of Waterloo）
时 间：2015年1月14日(星期四)下午15:00
地 点：统计研究院426教室（原附中计算机与控制工程学院办公楼四楼东侧)
摘 要: Times of disease progression are interval-censored when progression status is only known at a series of assessment times.This situation arises routinely in clinical trials and cohort studies when events of interest are only detectable upon imaging, based on blood tests, or upon careful clinical examination. Interest lies in selecting important prognostic biomarkers from a large set of candidates and assessing the accuracy of this predictive model when disease progression status is only known at irregularly spaced and individual-specific assessment times. We propose a method for penalized regression (e.g. LASSO, adaptive LASSO and SCAD) to handle interval-censored time of disease progression. A flexible parametric model with piecewise constant baseline hazard function is adopted and an expectation-maximization algorithm is described which is empirically shown to perform well. To evaluate the predictive performance, inverse probability weighted (IPW) estimators of the mean prediction error and the area under the receiver operating characteristic curve are developed and evaluated. The weights are estimated from a multi-state model
which jointly considers the event process of interest along with the inspection and censoring processes.
欢迎广大师生参加！
统计研究院
2015年12月31日