From Data to Knowledge in Secondary Health Care Databases

Bettencourt-Silva, Joao (2014) From Data to Knowledge in Secondary Health Care Databases. Doctoral thesis, University of East Anglia.

[img]
Preview
PDF
Download (15MB) | Preview

Abstract

The advent of big data in health care is a topic receiving increasing
attention worldwide. In the UK, over the last decade, the National
Health Service (NHS) programme for Information Technology
has boosted big data by introducing electronic infrastructures in hospitals
and GP practices across the country. This ever growing amount of
data promises to expand our understanding of the services, processes
and research. Potential bene�ts include reducing costs, optimisation
of services, knowledge discovery, and patient-centred predictive modelling.
This thesis will explore the above by studying over ten years
worth of electronic data and systems in a hospital treating over 750
thousand patients a year.
The hospital's information systems store routinely collected data, used
primarily by health practitioners to support and improve patient care.
This raw data is recorded on several di�erent systems but rarely linked
or analysed. This thesis explores the secondary uses of such data by
undertaking two case studies, one on prostate cancer and another on
stroke. The journey from data to knowledge is made in each of the
studies by traversing critical steps: data retrieval, linkage, integration,
preparation, mining and analysis. Throughout, novel methods
and computational techniques are introduced and the value of routinely
collected data is assessed. In particular, this thesis discusses
in detail the methodological aspects of developing clinical data warehouses
from routine heterogeneous data and it introduces methods to
model, visualise and analyse the journeys that patients take through
care. This work has provided lessons in hospital IT provision, integration,
visualisation and analytics of complex electronic patient records
and databases and has enabled the use of raw routine data for management
decision making and clinical research in both case studies.

Item Type: Thesis (Doctoral)
Faculty \ School: Faculty of Science > School of Computing Sciences
Depositing User: Mia Reeves
Date Deposited: 30 Jun 2015 12:26
Last Modified: 30 Jun 2015 12:26
URI: https://ueaeprints.uea.ac.uk/id/eprint/53416
DOI:

Actions (login required)

View Item View Item