Date of Award

2024

Document Type

Master Thesis

Degree Name

Master of Science

Department

Computing (Business, Computing & Humanities)

First Advisor

Andrew Shields

Second Advisor

Dr Pat Doody

Abstract

National cancer registries play an important role in understanding the patterns and trends of cancer diseases, prevention methods, and treatment management. However, issues arising from data silos and an inability to link heterogeneous data sources have led to problems within the healthcare industry. This became apparent in Ireland in 2018 when Dr Scally’s scoping inquiry into the CervicalCheck controversy highlighted patient safety issues relating to disjointed data and an inability to link datasets.

Worldwide, semantic technologies are being utilised for their ability to link diverse data and enhance analytical capabilities. Therefore, this thesis will examine the use of a knowledge graph to solve the data management issues currently facing the Irish cancer registry system. An ontology will be developed to link Irish cancer registry data and incorporate standardised medical classifications. A knowledge graph will then be created using the Neo4j graph database system, and a comparison of querying capabilities will show the benefits of a knowledge graph system over the existing relational database management systems used today.

The Irish cancer registry data will be combined with external sources showing the ability to seamlessly integrate heterogeneous data across clinical and non-clinical organisations. Finally, data visualisations generated from the linked data within the knowledge graph will show how this method can present unique data insights and offer a paradigm for harmonising varying sources of data to expand today’s data visualisation capabilities.

Access Level

info:eu-repo/semantics/openAccess

Download

Share

COinS