Research

This project will use data from MR imaging and duplex ultrasonography to develop personalised in silico and in vitro hemodynamic models of the carotid artery bifurcation to predict stroke in patients with carotid artery stenosis.

This project will use 4D ultrasound images of Abdominal Aortic Aneurysms (AAAs) to detect intraluminal thrombus (ILT) motion between systole and diastole through speckle tracking. These data, combined with an inverse method, will help estimate locally varying mechanical properties of patient-specific ILTs and will better evaluate biomechanical markers of rupture risk in patients with AAAs.

This project will develop personalised in silico and in vitro haemodynamic models to predict graft thrombosis in haemodialysis patients. Anatomical and functional data will be collected to create in silico and in vitro microfluidic models, providing a tool for vascular access surveillance to predict patient-specific stenosis severity, which, if left untreated, can lead to graft thrombosis.

This project will develop personalised in silico and in vitro haemodynamic models of venous thrombosis to predict residual vein obstruction, recurrence, and post-thrombotic syndrome (PTS) in patients with deep vein thrombosis (DVT), providing a patient-specific risk prediction tool for recurrent DVT.

This project will develop predictive computational models to capture initial thrombus formation and the heterogeneous porous structure of microthrombi. These models will incorporate fluid mechanical effects from the outset, calculating structural stresses and agonist transport. Model parameters and validation data will be obtained from experimental imaging.

In this project, microfluidic in vitro experiments will be used to analyse platelet aggregate formation under different shear conditions to inform the development of predictive computational models and to provide fundamental knowledge of platelet activation dynamics, the involvement of von Willebrand factor, and the formation of distinct platelet subpopulations during aggregation.

This project will investigate the role of clot contraction in thrombosis and thromboembolism and develop a hybrid assay that combines a simplified in vitro blood test with an in silico model of thrombus growth, enabling assessment of thrombosis risk in medical devices.

The project will develop a systematic approach using a matrix of samples with varied surface properties but identical base material. These samples will undergo rigorous characterisation to isolate specific surface attributes, followed by targeted assays of blood interactions. This approach aims to reveal the distinct contributions of each property to blood-material interactions, providing a clearer understanding of their roles in thrombus formation.

This project will develop predictive in silico models of thrombosis in the false lumen of Type-B Aortic Dissections using Challenge-Based Learning (CBL) principles, and explore the correlation between haemodynamic markers and thrombotic outcomes within a patient cohort.

This project aims to simulate and study thrombosis in a well-controlled flow environment in the lab. The challenge is to develop microfluidic systems that replicate human vascular conditions, to create in vitro models that mimic individual patient blood flows and clotting dynamics, and to visualise and quantify these processes to support a data-driven approach to modelling thrombosis and patient risk.

This project aims to analyse the fundamental physical principles that underlie the development of computational models at various scales for thrombus dynamics. By integrating principles and primitives from the cellular, thrombus, and circulatory levels, the project will create a unified multi-scale physics model of thrombus dynamics.

This project aims to analyse and integrate multimodal data to develop a foundational model for thrombosis. It will focus on integrating physics-based data obtained from computational experiments with image-based and clinical measurements. Integrating physics-based data is expected to produce a more robust and accurate foundational AI model for thrombosis.

This project will develop a model of lower limb haemodynamics that includes the function of the muscle pump and the action of respiration, both separately and together; both mechanisms are expected to influence venous haemodynamics in the lower limb, with the calf muscle acting more locally and respiration more globally; it is expected that the observed changes in lower limb haemodynamics will be significant enough to influence the coagulation mechanisms involved in deep vein thrombosis (DVT) development.

The project aims to develop a Verification, Validation, and Uncertainty Quantification (VVUQ) framework for multi-scale modelling. It seeks to improve model reliability by identifying key parameters and quantifying uncertainties across various scales. The project will establish an optimised workflow to transfer findings from Uncertainty Quantification/Sensitivity Analysis (UQ/SA) experiments directly into model refinement processes.

This project will determine the impact of clot contraction on thrombus structure and function. It aims to demonstrate the effects of clot contraction in in vivo models of pulmonary embolism and to develop in silico models of clot contraction. Clot structure will be assessed using Scanning Electron Microscopy (SEM) and confocal microscopy, while clot stability will be studied using magnetic tweezer microrheology and Atomic Force Microscopy (AFM) force-pulling. Experimental findings will be integrated with in silico models and examined under flow using microfluidic models.

The project aims to design MR-based imaging biomarkers and investigate thrombus formation using Magnetic Resonance Imaging (MRI), Photon-Counting Computed Tomography (PCCT), and thrombus material. It will improve in vivo characterisation of thrombus composition and develop a dedicated thrombus imaging protocol for clinical routine. The project will also correlate findings with the growth behaviour of Abdominal Aortic Aneurysm (AAA) and aortic dissection over time and under best medical treatment.

This project aims to identify improved valve designs to reduce the risk of bioprosthetic valve thrombosis. To achieve this, our existing high-fidelity fluid-structure interaction solver must be integrated with advanced models of blood damage and thrombosis to predict thrombosis risk. The long-term vision for personalised medicine is to create patient-specific valve designs that can be manufactured on demand (for example, by 3D printing) for individual patients, taking into account patient anatomy, blood coagulability, comorbidities, and other relevant factors.

This project aims to define a new treatment strategy for pulmonary embolism using catheter-based infusion of thrombolytic drugs into the pulmonary circulation and to improve understanding of thrombus properties. To this end, it will investigate the use of an existing device, currently used for the diagnosis and treatment of coronary microvascular obstruction, in a similar pathology in the lung.