Masters Theses

Date of Award

8-1994

Degree Type

Thesis

Degree Name

Master of Science

Major

Nuclear Engineering

Major Professor

Peter G. Groer

Committee Members

Laurence Miller, Michael Fry

Abstract

A Bayesian statistical method was developed to estimate the parameters for the following survival model characterized by the probability density f(t,d)=&alpha&lambda&sup&alpha&iota&sup&alpha-1&e&sup&rho&betade&sup-(&lambda&iota)&sup&alphae&rho&betad Computer programs were developed in "Mathematica" (10) to calculate the posterior densities for the model parameters &alpha, &beta, &lambda, and &rho. The data used in the analyses came from BALB/c female mice exposed to ¹37Cs gamma radiation (5,6). The doses ranged from 0 to 2 Gy with dose rates from 0.083 Gy/day to 0.4 Gy/min (5,6). The data were generated at the Biology division of Oak Ridge National Lab (ORNL) and supplied by ORNL and the National Radiobiology Archives. The parameter &rho describes the effect of dose-rate on the induction of cancer in mice. The derived posterior density for p describes the remaining uncertainty about the parameter after updating. This thesis describes the first direct estimation of a dose-rate effectiveness factor (DREF) using the techniques of Bayesian parameter estimation and gives the first quantitative description of the uncertainty of this factor in terms of a probability density.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS