Doctor of Philosophy (Ph.D.)
Degree Granting Department
Mathematics and Statistics
Chris P. Tsokos, Ph.D.
Lu Lu, Ph.D.
Kandethody Ramachandran, Ph.D.
Dan Shen, Ph.D.
stomach cancer, parametric analysis, tumor size, quantile regression, survival analysis
The objective of this study is to address some important questions associated with stomach cancer patients using the data from the Surveillance Epidemiology and End Results (SEER) program of the United States. To better understand the behavior of stomach cancer, we first perform parametric analysis for each patient group (white male, white female, African American male, African American female, other male and female) to identify the probability distribution function which can best characterize the behavior of the malignant stomach tumor sizes. We evaluate the effects of patients’ age, gender and race on the malignant stomach tumor sizes by developing quantile regression models, which gives us a better understanding of the behavior of the malignant stomach tumors.
We also proposed statistical models with respect to patients’ malignant stomach tumor size as a function of age for different races and gender group, respectively. The proposed models were evaluated to attest their prediction quality. Furthermore, we have identified the rate of change of the malignant tumor size as a function of age, for gender and race.
We evaluated the routine treatment of stomach cancer using parametric and nonparametric survival analysis. We have found that stomach cancer patients who receive surgery with radiation together have a better survival probability than the patients who receive only radiation. We performed decision tree analysis to assist the physician in recommending to his patients the most effective treatment that is a function of their characteristics.
Scholar Commons Citation
Gao, Chao, "Statistical Analysis and Modeling of Stomach Cancer Data" (2017). Graduate Theses and Dissertations.