Mapping an acute knee injury database to an ontology to support future epidemiology studies

Aleksandra I Nacheva

Supervised by Irena Spasic, Moderated by Jianhua Shao

As part of ongoing collaboration between the School of Computer Science & Informatics and the School of Healthcare Sciences, we have developed an extensive ontology to formally describe concepts related to diagnosis and treatment of knee conditions (approximately 2,000 terms). While there are many biomedical ontologies, this particular ontology is the first of its kind in physiotherapy and as such is breaking new ground in healthcare. In particular, it can be used to support coded data collection in support of epidemiology studies. However, there is a need to convert the legacy data collected over the course of 12 years in the Acute Knee Screening Clinic at Cardiff and Vale University Health Board. The database contains data about injury types and mechanisms, investigations and interventions, all concepts currently described in our ontology. The purpose of this project is to semi-automatically re-structure and code the given database. This would be supported by actively querying the database in order to exploit any observed regularities to implement the mapping rules from free text entries to concepts in the ontology. The student would expand their data analysis and programming skills. They will learn about using ontologies to formally represent knowledge and use them to support unambiguous data interpretation by both human users and computer applications. They would also learn to actively collaborate with professionals from other disciplines and in particular develop knowledge elicitation skills, which are a common bottleneck in interdisciplinary studies.

Initial Plan

Final Report