Thesis

Computational haemodynamics in vascular networks : multi-phase cell-free layer modelling in the microvasculature, testicular microcirculation, and blood flow in the pulmonary bifurcation

Downloadable Content

Download PDF
Creator
Rights statement
Awarding institution
  • University of Strathclyde
Date of award
  • 2026
Thesis identifier
  • T17652
Person Identifier (Local)
  • 202182221
Qualification Level
Qualification Name
Department, School or Faculty
Abstract
  • Small vessel disease is recognised as a major contributor to serious clinical outcomes, including ischaemia, stroke, cardiac insufficiency, renal dysfunction and organ atrophy, due to the associated disruption in the microvascular blood flow. The precise cause-and-effect relationship between localised haemodynamics and clinical outcomes in such diseases remains poorly established. For that, it is important to understand the underlying mechanisms of blood flow in microvascular networks. Utilising Computational Fluid Dynamics (CFD) methods, which help quantify haemodynamic parameters not easily measured in vivo, this thesis investigates simplified and anatomically correct vessel geometries at various scales. A key novelty of this work is the development of a novel multi-phase, multicomponent continuum-based model using the Eulerian Multiphase (EMP) method, which treats blood as a suspension of red blood cells (RBCs) in plasma. With this, we investigate the behaviour of the cell-free layer (CFL) in arteriolar microvessels, a critical plasma-rich lubricating layer adjacent to the vascular wall that is fundamentally disrupted at bifurcations. This disruption creates a lasting asymmetric flow profile in the daughter vessels that governs overall perfusion in a microvascular network. The model was validated using a CFL value specific to the haematocrit and diameter of a vessel, to allow assessment of its rheological predictions. The model’s predictions were compared with established experimental findings to test its ability to reproduce fundamental non-Newtonian blood characteristics, including shear-thinning behaviour and the non-linear viscosity–haematocrit relationship. The Power-law, Casson, and Carreau–Yasuda models were used as benchmarks, confirming that the EMP approach robustly captured the shear-dependent viscosity of blood and its haematocrit dependence. Parametric investigations then evaluated the model’s robustness, and a range of physiological and geometrical conditions were studied. These conditions included variations in the feeding haematocrit, inflow rate, biased outflow, and bifurcation architecture. The simulations reproduced well-known haemodynamic phenomena in arterioles, such as the haematocrit partitioning (the Zweifach–Fung effect), asymmetric CFL profiles, and the downstream recovery of flow symmetry. These results provided strong evidence that the EMP model not only reproduces physiologically observed phenomena but also offers a mechanistic insight into their haemodynamic origins. The framework was then extended to realistic arteriolar microvessel network reconstructions. The multi-phase, multi-component simulations were conducted in testicular bifurcated microvessels with non-planar, tortuous paths. These studies revealed how CFL formation and plasma skimming manifest in physiologically relevant geometries. This work bridges the gap between idealised bifurcations and real vascular architecture, demonstrating the model’s capacity to capture the effects of complex geometry on local haemodynamics. In network scale, the multi-phase, multi-component model was successfully applied to the testicular atrophic network (48 µm). However, its application to the control network (89 µm), which features more extensive branching, was omitted due to the significant meshing challenges, which would require mesh optimisation. The singlephase model was instead used, revealing distinct haemodynamic patterns between control and atrophic morphologies, which translating into functional deficits in perfusion and resistance. To establish a clear contrast with the micro-scale, a separate analysis was performed on a large vessel bifurcation, the pulmonary arteries, where inertial forces are stronger and the CFL effects are negligible. By establishing this macro-scale simulation, the unique and dominant role of multi-phase, multicomponent phenomena in microcirculation becomes clearer. Successfully bridging these distinct vascular scales underscores the value of the computational methods employed throughout this thesis.
Advisor / supervisor
  • Kazakidi, Asimina
  • Wu, Junxi
Resource Type
DOI

Relations

Items

Thumbnail Title Date Uploaded Visibility Actions
file details PDF of thesis T17652 2026-04-15 Public Download