Quick Search:

Development and evaluation of a semantic-based access framework for health record linkage

Lu, Yang ORCID: https://orcid.org/0000-0002-0583-2688 (2018) Development and evaluation of a semantic-based access framework for health record linkage. Doctoral thesis, University of Melbourne.

Full text not available from this repository.

Abstract

Record linkage is the task of finding records in data sources that refer to the same entity across those different resources. In the health domain, record linkage is commonly used for identifying records belonging to the same individual (patient) and augmenting the patient data with additional information obtained from those data sources. A multitude of centralised record linkage centres have been established in the health domain in Australia and indeed internationally, however the vast majority of health data resides in heterogeneous primary and secondary care settings and is not directly accessible for linkage or for external research purposes.

In this context, the need to support improved record linkage is clear. Ideally this should leverage existing information and processes as much as possible from distributed and autonomous stakeholders. This includes data resources as well as associated organisational security policies. However existing security policies are predominantly inflexible and typically do not lend themselves for direct re-use. Furthermore, a common demand in record linkage is for privacy and confidentiality. In the modern data-rich world, ensuring privacy and confidentiality of data is increasingly challenging since seemingly innocuous and non-identifying health data released following a linkage process, e.g. ethnicity, can be reused to erode privacy and potentially re-identify individuals.

To address this, this thesis presents a three-layer semantically driven access control framework comprising a range of components essential for flexible and secure data linkage. Specifically, this framework leverages pre-existing security information and policies to support more flexible access to data required for decentralised data linkage scenarios. It supports reasoning on privacy issues that can arise including the risks associated with direct data linkages and potential inclusion of external data resources when data is to be linked and released. The framework also balances data privacy demands with the subsequent utility of the data.

The framework is applied in a range of health-related settings including targeted international disease registries and use of population-wide data sets for data linkage where privacy concerns are paramount.

Item Type: Thesis (Doctoral)
Status: Unpublished
Subjects: H Social Sciences > HB Economic Theory > HB135-147 Mathematical economics. Quantitative methods
Q Science > Q Science (General)
Q Science > Q Science (General) > Q325 Machine learning
Q Science > QA Mathematics
Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Q Science > QA Mathematics > QA76 Computer software
Q Science > QA Mathematics > QA76.9.H85 Human-Computer Interaction; Virtual Reality; Mixed Reality; Augmented Reality ; Extended Reality
R Medicine > RA Public aspects of medicine > RA0421 Public health. Hygiene. Preventive Medicine
Z Bibliography. Library Science. Information Resources > ZA Information resources
Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4050 Electronic information resources
Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4450 Databases
School/Department: School of Science, Technology and Health
URI: http://ray.yorksj.ac.uk/id/eprint/6004

University Staff: Request a correction | RaY Editors: Update this record