Use of Machine-Learning Software to Categorize Oncotype DX Data Public Deposited

Downloadable Content

Download PDF
Last Modified
  • April 10, 2022
Creator
  • Gomez,Keven
    • Affiliation: Eshelman School of Pharmacy
Abstract
  • This study looked into the utilization of machine-learning software known as the Clinical Annotation Research Kit or CLARK to abstract and structure Oncotype DX testing results, a genomic test result that is often reported in text-heavy unstructured clinic notes, into a format that can be merged with larger structured databases. Oncotype DX testing results were abstracted for all the patients and each were labeled both manually and through CLARK as being low, intermediate, high or no score based on their risk level. The labeling accuracy of CLARK was compared to manual abstraction.
Date of publication
DOI
Resource type
Rights statement
  • In Copyright
Advisor
  • Roberts, Megan
    • Affiliation: Eshelman School of Pharmacy, Division of Pharmaceutical Outcomes and Policy
  • Kinlaw ,Alan
    • Affiliation: Eshelman School of Pharmacy, Division of Pharmaceutical Outcomes and Policy
Degree
  • Doctor of Pharmacy
Graduation year
  • 2022
Language
  • English
Parents:

This work has no parents.

Items