【電機系書報討論演講訊息】

書報討論網頁：http://www.ee.nthu.edu.tw/~ee591000/

演講公告網址：/p/403-1175-2629-1.php

題   目：Analysis and Optimization of a Life Table Embedded Disease Progression Model

講   員：陳政輝 教授 (國立政治大學)

時      間：2013年11月22日(星期五) 下午2時20分 (2時10分入場)

地      點：資電館B01演講廳

abstract：

Progressions of chronic diseases can be modeled as Markov processes. Frequently, the model parameters are concluded based on distinct clinic studies, such as meta-analysis, due to the difficulty of observing the entire progression process in one clinic study. Though this piece-by-piece approach depicts a global picture to the disease progression, it could lead to some unrealistic results under in-depth analysis. For instance, patients may have longer life expectancy than the general populations. This contradicted result generally arises from that the effect of mortality of natural death is not well observed or considered in these short-term clinic studies. It has been proposed that the life table of the general population can be introduced into the Markov model to resolve this problem. This method provides a more realistic approximation to the life expectancy of patients’ population. However, the effect caused by the inclusion of a life table into a Markov model has not been thoroughly investigated. In this work, a rigorous mathematical treatment to this effect is provided. Specifically, the difference between life expectancies estimated based on models with and without life table is analyzed. Explicit formulae of an upper and a low bound to the estimation difference are offered. These bounds clearly indicate how the morality of life table affects the estimation difference of these two models. They provide insightful information to how clinic studies should been conducted when the mortality due to natural death is not negligible. Meanwhile, the problem to optimally model disease progression processes under such a circumstance can be addressed based on this work. An example taken from chronic hepatitis B virus infection is presented to demonstrate the theoretical derivations. Further extensions to this work will also be discussed and considered.

Biography

Professor Jeng-Huei Chen received his bachelor’s degree from National Tsing-Hua University. He was awarded his Ph.D. degree from the School of Electrical and Computer Engineering, Cornell University. He joined the Department of Mathematical Sciences, National Chengchi University subsequently and is currently an assistant professor. His major research is optimization methods and mathematical modeling with focus on chronic disease progression (chronic hepatitis B virus infection and diabetes) and cancer treatments. He is a member of INFORMS.

~歡  迎  聽  講~