Computational modelling of diffusion tensor cardiovascular magnetic resonance: effects of membrane permeability, perfusion and strain
File(s)
Author(s)
Alemany, Ignasi
Type
Thesis
Abstract
Diffusion Tensor Cardiovascular Magnetic Resonance (DT-CMR) is a histology-validated, non-invasive imaging technique that assesses the average microstructure of the myocardium within a specific voxel region (usually around 2 × 2 × 8 mm³) by analyzing the direction and magnitude of the self-diffusion of water molecules (Brownian motion). However, due to motion artifacts, confounding factors, and complex tissue microstructure, the sensitivity of DT-CMR tensor parameters to the microstructural characteristics—as well as the link between the tensor parameters and such characteristics—remains poorly understood. Numerical phantoms offer a controlled framework to elucidate this link by allowing the independent manipulation of variables in a physically well-defined model. This thesis aims to enhance DT-CMR simulations by integrating several key biophysical processes, thereby improving the understanding of tensor sensitivity across in-vivo sequences of clinical relevance. Specifically, this work introduces several key methodological advances for Monte Carlo Random Walk (MCRW) simulations of DT-CMR: the implementation of a hybrid transit model that accurately and efficiently simulates membrane permeability; an enhanced perfusion model incorporating both temporal variations in capillary flow and inter-capillary velocity dispersion; and a validated strain implementation. Additionally, a sheetlet packer framework is developed and validated against histology data, enabling the efficient generation of realistic cardiac microstructure.
Version
Open Access
Date Issued
2024-11-03
Date Awarded
2025-03-01
License URL
Advisor
J. Doorly, Denis
D. Scott, Andrew
Publisher Department
Department of Aeronautics
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)