Year

2012

Degree Name

Master of Information and Communication Technology - Research

Department

School of Information Systems and Technology

Abstract

The development of medical knowledge has always proceeded along pathways that relate to questions arising from expert domain specialty knowledge, practice and work flow. While medical knowledge advances as a whole, there are independent and specific knowledge streams for each specialty.

Radiation Oncology is a specialty domain, therefore one can expect that its ontology will reflect and be defined by its clinical terms and clinical work ow which are both specific and interconnected.

The assertion of this thesis is that this particular medical domain specialty demonstrates clearly that knowledge derived in the clinical management of a patient, the processes used in clinical trials to develop and report new knowledge to be used in the clinical management of the patient, and the recommendations found in clinical guidelines that inform routine patient management, are all views of the same domain specific knowledge structure.

The process of Content Analysis was adapted and applied manually to an objective corpus of Radiation Oncology literature to establish the semantic entities underlying the knowledge being communicated to specialist radiation oncologists. These semantic entities were grouped in a hierarchy based on the work ow relevant to the expert domain. The hierarchical structures were then coded into a schema using an XML format to form a "Clinical Knowledge Mark up Language" (CKML).

The applicability of this CKML schema is subsequently demonstrated as relevant by its application to four clinical scenarios: description of a normal clinical patient with locally advanced oropharyngeal cancer, description of a clinical trial protocol for which patients with locally advanced oropharyngeal cancer may be eligible, specification of a clinical trial result from this same trial, and lastly as a description of clinical guidelines describing the management of patients with locally advanced oropharyngeal cancer.

Share

COinS