UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Enhancing magnetic resonance imaging with computational fluid dynamics

Annio, Giacomo; (2020) Enhancing magnetic resonance imaging with computational fluid dynamics. Doctoral thesis (Ph.D), UCL (University College London). Green open access

[thumbnail of Enhancing_Magnetic_Resonance_Imaging_with_CFD_G.Annio.pdf]
Enhancing_Magnetic_Resonance_Imaging_with_CFD_G.Annio.pdf - Accepted Version

Download (32MB) | Preview


Quantitative assessment of haemodynamics has been utilised for better understanding of cardiac function and assisting diagnostics of cardiovascular diseases. To study haemodynamics, magnetic resonance imaging (MRI) and computational fluid dynamics (CFD) are widely used because of their non-invasive nature. It has been demonstrated that the two approaches are complementary to each other with their own advantages and limitations. Four dimensional cardiovascular magnetic resonance (4D Flow CMR) imaging enables direct measurement of blood flow velocity in vivo while spatial and temporal resolutions as well as region of image acquisition are limited to achieve a detailed assessment of the haemodynamics. CFD, on the other hand, is a powerful tool that has the potential to expand the image-obtained velocity fields with some problem-specific assumptions such as rigid arterial walls. We suggest a novel approach in which 4D Flow CMR and CFD are integrated synergistically in order to obtain an enhanced 4D Flow CMRI (EMRI). The enhancement will consist in overcoming the spatial-resolution limitations of the original 4D Flow CMRI, which will enable more accurate quantification of flow dependent bio-mechanical quantities (e.g. endothelial shear stress) as well as non-invasive estimation of blood pressure. At the same time, it will reduce a number of assumptions in conventional haemodynamic CFD such as in/outflow conditions including the effect of valves, the impact of patient-specific vessel wall motion and the effect of the surrounding tissues. The approach was first tested on a 2D portion of a pipe, to understand the behaviour of the parameters of the model in this novel framework. Afterwards the methodology was tested on patient specific data, to apply it to the analysis of blood flow in a patient specific human aorta, in 2D. The outcomes of EMRI are assessed by comparing the computed velocities with the 4D Flow CMR one. A fundamental step to allow the translation to clinics of this methodology was the validation. The study was performed on an idealised-simplified model of the human aortic arch – a U bend – with a sinusoidal inflow applied by a pump. Firstly, phase resolved particle image velocimetry (PIV) (an experimental technique enables high spatial-temporal resolution) was performed in 5 different time points of the pump cycle, using a blood alike fluid with the same refractive index matched of the clear silicon phantom, and seeded with silver coated hollow glass spheres. Real time 4D Flow CMR was then performed on the phantom with MRI. Lastly using the pump flow rate and the phantom geometry, a computation of the flow through the U bend was conducted using Ansys CFX. The flow patterns obtained from the 3 methods were compared in the middle plane of the phantom. The methodology was then applied to study a patient specific aorta in 3D, and retrieve flow patterns and flow dependent parameters. Finally, the validated methodology was applied to study atherogenesis, and in particular to investigate the relation between EMRI retrieved flow quantities (e.g. wall shear stress (WSS)) and temperature heterogeneity. A carotid artery phantom was realised and studied with CFD, MRT and EMRI. All the results demonstrate that EMRI preserves flow structures while correcting for experimental noise. Therefore it can provide better insights of the haemodynamics of cardiovascular problems, overcoming the limitations of 4D Flow CMR and CFD, even when considering a small region of interest. These findings were supported by the validation experiment that showed how EMRI retrieved flow patterns were much more consistent with the one measured with high resolution PIV, compensating for 4D Flow CMR errors. These findings lead to the application to the atherogenesis problem, allowing higher resolution flow patterns, more suitable to be compared to the temperature distribution and highlighted how flow patterns exert an influence on the temperature distribution on the vessel wall. EMRI confirmed its potential to provide more accurate non-invasive estimation of flow derived and flow dependent quantities and become a novel diagnostic tool.

Type: Thesis (Doctoral)
Qualification: Ph.D
Title: Enhancing magnetic resonance imaging with computational fluid dynamics
Event: UCL
Open access status: An open access version is available from UCL Discovery
Language: English
Additional information: Copyright © The Author 2020. Original content in this thesis is licensed under the terms of the Creative Commons Attribution 4.0 International (CC BY 4.0) Licence (https://creativecommons.org/licenses/by/4.0/). Any third-party copyright material present remains the property of its respective owner(s) and is licensed under its existing terms. Access may initially be restricted at the author’s request.
UCL classification: UCL
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Mechanical Engineering
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > STEaPP
URI: https://discovery.ucl.ac.uk/id/eprint/10110638
Downloads since deposit
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item