Multimodal Modelling of Personalised Haemodynamics in Heart Failure with Preserved Ejection Fraction (HFpEF)

Project Overview

Heart failure affects millions, yet personalised treatments remain limited. This project develops advanced computational models to simulate the progression of patient-specific heart function in heart failure with preserved ejection fraction (HFpEF). It combines image-based fluid–structure interaction modelling and statistical emulation to improve diagnosis and enable more precise treatments.

What You Will Do

You'll work with clinical imaging data, such as magnetic resonance imaging (MRI) and computed tomography (CT), to build detailed patient-specific models that capture both heart mechanics and blood flow through fluid–structure interaction approaches. These models will be used to estimate important physiological properties, including myocardial stiffness and filling pressures, in a non-invasive way.

A key part of the project will involve creating virtual patient cohorts to represent different HFpEF phenotypes and using these to investigate how patients may respond to various treatments through in-silico studies. To make these simulations more efficient, you'll explore the use of statistical emulators and machine learning methods, enabling faster predictions and supporting inverse modelling approaches for personalised parameter estimation.

Expected Outcomes

This project will produce novel computational tools to facilitate better diagnosis and tailored therapies for HFpEF patients. The development of efficient, patient-specific models and their validation through virtual cohorts is expected to pave the way for clinical applications.

Why It Matters

Heart failure is a major global health burden with limited personalised treatment options. By advancing computational haemodynamics modelling interfaced with clinical imaging, this research addresses critical gaps in understanding and managing HFpEF, ultimately aiming to improve patient outcomes and care strategies.