以轉換治癒模型分析區間設限資料 - 低壓減壓症之實例分析

在醫學研究中, 收集資料的過程通常持續一段時間, 在此情形下,資料多為區間設限資料。此類型資料在存活分析、社會科學、工業品管的研究中廣泛的被使用。再者, 這類資料中有些個體僅有極低的發生機會為有興趣的事件, 而這些個體即定義為治癒或不易感受性。我們建立混合半參數治癒模型, 藉此用來分析包含區間設限資料及不易感受性資料。本篇論文中, 我們目的是分析低壓減壓症資料。此資料研究重點為調查在低壓減壓的環境中發生疾病的風險。我們透過混合半參數治癒模型提出概似函數並使用EM 演算法進行迴歸參數的估計。將此方法應用在低壓減壓症資料, 並與其他模型與方法進行比較。

關鍵字

EM 演算法；區間設限資料；治癒模型；轉換模型

並列摘要

In a medical study, it takes a period of time to collect the event we are interested in and most of these data are collected by a time interval. The time interval (interval-censored) arises in many other investigations including sociological studies, reliability experiments and biological studies. In addition, the disease may be completely eliminated, that is a significant fraction of patients can be cured in a biomedical research. The cure rate represents a combination of cure proportion and survival model. Here, we construct a semi-parametric cured model to analysis the interval-censored data and cure data. In this thesis, our goal is to analysis the Hypobaric Decompression Sickness Data (Conkin et al., 1992). We investigate the risk of decompression sickness in hypobaric environments. We present the likelihood function form the model we proposed and estimate the parameters by the Expectation Maximization (EM) algorithm. Based on our proposed method, we compare with other methods by the Hypobaric Decompression Sickness Data.

並列關鍵字

EM algorithm ； interval censoring ； transformation model ； cure model

參考文獻

[5] Finkelstein, D. M. and Wolfe, R. A.(1985).A semiparametricmodel

[1] Agresti, A. (2002).Categorical Data Analysis. Wiley. New York.

for multivariate survival data with a surviving fraction. Journal of Multivariate Analysis, 80, 101-126.

[4] David G. Kleinbaum, Mitchel Klein(2005).Survival Analysis:

for regression analysis of interval-censored failure time data. Biometrics, 41, 933-945.

國際替代計量

以轉換治癒模型分析區間設限資料 - 低壓減壓症之實例分析

全文下載

主題瀏覽