Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.641038
Title: Personalised finite-element models using image registration in parametric space
Author: Silva Soto, Daniel Alejandro
ISNI:       0000 0004 5350 2872
Awarding Body: University of Sheffield
Current Institution: University of Sheffield
Date of Award: 2014
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
Heart failure (HF) is a chronic clinical condition in which the heart fails to pump enough blood to meet the metabolic needs of the body. Patients have reduced physical performance and can see their quality of life severely impaired; around 40-70% of patients diagnosed of HF die within the first year following diagnosis. It is underestimated that 900,000 people in the UK currently suffer from HF. HF has a big impact on the NHS, representing 1 million inpatient bed, 5% of all emergency medical admission to hospitals and costs 2% of the total NHS budget. The annual incidence of new diagnoses is reported as 93,000 people in England alone – and this figure is already increasing at a rate above that at which population is ageing [1]. Cardiac resynchronisation therapy (CRT) has become established as an effective solution to treat selected patients with HF. The research presented in this thesis has been conducted as part of a large EPSRC-Funded project on the theme of Grand Challenges in Heathcare, with co-investigators from King’s College London (KCL), Imperial College London, University College London (UCL) and the University of Sheffield. The aim is to develop and to apply modelling techniques to simulate ventricular mechanics and CRT therapy in patient cohorts from Guy’s Hospital (London) and from the Sheffield Teaching Hospitals Trust. This will lead to improved understanding of cardiac physiological behaviour and how diseases affect normal cardiac performance, and to improved therapy planning by allowing candidate interventions to be simulated before they are applied on patients. The clinical workflow within the hospital manages the patient through the processes of diagnosis, therapy planning and follow-up. The first part of this thesis focuses on the development of a formal process for the integration of a computational analysis workflow, including medical imaging, segmentation, model construction, model execution and analysis, into the clinical workflow. During the early stages of the project, as the analysis workflow was being compiled, a major bottle-neck was identified regarding the time required to build accurate, patient-specific geometrical meshes from the segmented images. The second part of this thesis focuses on the development of a novel approach based on the use of image registration to improve the process of construction of a high-quality personalised finite element mesh for an individual patient. Chapter 1 summarises the clinical context and introduces the tools and processes that are applied in this thesis. Chapter 2 describes the challenges and the implementation of a computational analysis workflow and its integration into a clinical environment. Chapter 3 describes the theoretical underpinnings of the image registration algorithm that has been developed to address the problem of construction of high-quality personalised meshes. The approach includes the use of regularisation terms that are designed to improve the mesh quality. The selection and implementation of the regularisation terms is discussed in detail in Chapter 4. Chapter 5 describes the application of the method to a series of test problems, whilst Chapter 6 describes the application to the patient cohort in the clinical study. Chapter 7 demonstrates that the method, developed for robust mesh construction, can readily be applied to determine boundary conditions for computational fluid dynamics (CFD) analysis. Chapter 8 provides a summary of the achievements of the thesis, together with suggestions for further work.
Supervisor: Hose, D. R. ; Wood, S. M. Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.641038  DOI: Not available
Share: