Date of Award
January 2023
Document Type
Thesis
Degree Name
Master of Science (MS)
Department
Biomedical Engineering
First Advisor
Sandeep Singhal
Abstract
Breast cancer biomarkers have great potential in providing clinicians more individualized information about the composition and outcomes of a patient’s breast cancer. However, many breast cancer biomarkers have not been evaluated on a large scale or in groups of patients with diverse characteristics, leading to difficulty in their translation to having an impact on patients. In this study, we compile a large, pooled breast cancer patient data cohort and evaluate breast cancer biomarkers on patients with diverse characteristics. Biomarkers are found to have varying expression patterns within the different breast cancer subtypes, validating the need to evaluate biomarkers on patient populations with diverse backgrounds, subtypes, and other breast cancer characteristics. As expected, ESR1, an estrogen receptor biomarker, showed significant increased expression in the Luminal A and Luminal B subtypes for this dataset. The large, pooled cohort developed in this study has future potential in many areas of breast cancer research.
Recommended Citation
Scheafer, Kalli, "Machine Learning And Data Mining To Validate The Prognostics And Predictive Breast Cancer Biomarkers On A Large Racially Diverse Population" (2023). Theses and Dissertations. 5265.
https://commons.und.edu/theses/5265