SU‐E‐T‐23: Ontological Representation of Radiation Treatment Guidelines

Published

Conference Paper

Purpose: To develop a Radiation Treatment Guidelines Ontology (RTGO) that represents the concepts, knowledge, and guidelines in radiation therapy using a formal and computable framework. This ontology will serve as the foundation of a decision support system for radiation oncologists. Method and Materials: The ontology framework chosen to represent RTGO is the Web Ontology Language (OWL) (www.w3.org/TR/owl‐guide). The OWL is rooted in the rigorous foundation of Description Logic and represents the latest development in ontology languages. The Protégé‐OWL ontology editor and knowledge‐base framework (http://protégé.stanford.edu) is used to build the ontology. The ontology is developed in a multi‐layered, modular, and iterative fashion. Upper ontologies are used to provide a sound structure. Concepts from standard ontologies are adopted whenever possible. Salient concepts and relationships are incorporated into the ontology through multiple iterations of extraction, modeling, implementation, and evaluation. Each iteration examines increasingly detailed concepts and relationships. The concepts are asserted in a strict subsumption hierarchy and are carefully formulated to be at a proper level of abstraction consistent with anticipated questions addressed by the ontology. Major ontology resources used in this project include National Center Biomedical Ontology's Bioportal (bioportal.bioontology.org) and National Library of Medicine's Universal Medical Language System (UMLS) (www.nlm.nih.gov/research/umls). Major ontologies consulted include FMA for anatomical concepts, SNOMED for clinical findings and symptoms, and Semantic Network for relationships. Results: The initial focus is on modeling the concepts and knowledge summarized in the Quantitative Analysis of Normal Tissue Effects in the Clinic (QUANTEC) papers. Starting with the organ specific papers on bladder and rectum, the ontology has so far captured 171 concepts, 80 relationships, and 21 individuals. The development is proceeding to include other organs in the QUANTEC literature. Conclusion: Early results demonstrate that major concepts in QUANTEC papers regarding dose volume effects can be successfully captured in an ontological framework. Duke University has a Master Research Agreement with Varian Medical Systems, INC. © 2011, American Association of Physicists in Medicine. All rights reserved.

Full Text

Duke Authors

Cited Authors

  • ge, Y; Minta, T; Kirkpatrick, J; Yuan, L; wu, Q

Published Date

  • January 1, 2011

Published In

Volume / Issue

  • 38 / 6

Start / End Page

  • 3490 -

International Standard Serial Number (ISSN)

  • 0094-2405

Digital Object Identifier (DOI)

  • 10.1118/1.3611973

Citation Source

  • Scopus