University of British Columbia
A computerised ontology to represent complex nursing knowledge for indicators of localised wound infection in chronic wounds
Abstract
dc:descriptionBACKGROUND: Managing wounds is a complex process carried out largely by nurses. Wound infection is a significant complication associated with increased risk of pain, delayed healing, amputation, sepsis, and death. Computerised ontologies, which are conceptual models of concepts, terms, and relationships represented in a computable form, could be positioned to assist clinicians in early identification of wound infection. The purpose of this dissertation was to develop a computerised ontology explicating nursing knowledge of localised wound infection identification in chronic wounds. METHODS: A three-phase qualitative interpretive description study was used. In Phase I, nurses were interviewed to conceptualize their real-world knowledge of identifying localised wound infection. In Phase II, expert wound care nurses were asked to examine the results from Phase I and a list of wound infection indicators from the literature. In Phase III, data from Phases I and II and the literature were synthesized to develop a conceptual model of concepts, terms, and relationships, which were subsequently used to build the computerised ontology in Protégé. RESULTS: In Phase I, a set of 8 high-level and 29 wound infection concepts, 101 synonym terms, and three themes were identified. In Phase II, the expert review resulted in retaining 25 concepts, adding 22 concepts, removing 4 concepts, adding synonym terms to 51 concepts, and removing synonym terms from 9 concepts. In Phase III, four conceptual models were identified: 1) relationship concepts, 2) highest and high-level class concepts, 3) infection indicator class concepts, and 4) assessment conclusion class concepts. This resulted in 52 highest and high-level concepts, 38 wound infection concepts, 160 synonym terms, and 3 relationship properties. The conceptual models were used to map concepts into Protégé. The final computerised ontology included 310 axioms, 91 classes, 159 sub-class axioms, 48 annotations, and 3 object properties. CONCLUSION: This research was able to demonstrate that rigorous qualitative methods can be used to identify complex nursing knowledge and model that knowledge in a computerised ontology. Future research should include further evaluation, expansion, and use of this ontology in advanced technology applications, such as artificial intelligence, to support and advance nursing practice.
Degree
thesis:*- Name thesis:degree_name
- Doctor of Philosophy - PhD
- Level thesis:degree_level
- doctoral
- Discipline thesis:degree_discipline
- Nursing
- Grantor dc:publisher
- University of British Columbia
- Year dc:date
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Block, Lorraine Joy
- Contributors dc:contributor
-
- Currie, Leanne M.
- Wong, Sabrina T.
Rights
dc:rights- Statement dc:rights
-
- Attribution-NonCommercial-NoDerivatives 4.0 International
- Language dc:language
- eng
Identifiers
dc:identifier.*- Handle dc:identifier
- http://hdl.handle.net/2429/89043