Computational Modelling

Our research in computational modelling focuses on the following main topics; clinical decision support, improved understanding of (patho)physiology and novel computational techniques.

On

Clinical decision support

Clinicians are starting to embrace computational models within the imaging modalities they use, such as CT and MRI. We have developed a system called VIRTUheart™ which takes invasive angiographic images, constructs a 3D model of the vessel, and applies computational fluid dynamics (CFD) to calculate the blood flow through diseased segments. This is the basis of much of our research on blood flow (using a ‘virtual’ fractional flow reserve, vFFR, approach) in patients with coronary artery disease. We seek to use this system to answer many relevant clinical questions, such as: Can we predict the results of stenting? What is the effect of limitation of flow in multi-vessel disease? Can we model the effects of treatment in patients’ everyday life? What about branches? And so on. The cover image above is of a coronary artery stenosis and the colour map denotes the calculated limitation of flow.

Valvular Heart Disease currently affects 2.5% of the population. It is overwhelmingly a disease of the elderly and as a consequence is on the rise. It is dominated by two conditions, Aortic Stenosis and Mitral Regurgitation, both of which are associated with significant morbidity and mortality, yet which pose a truly demanding challenge for treatment optimisation. The timing and nature of interventional treatment is crucial in valve disease, but it remains a major challenge in current clinical practice; in the EU more than 70,000 valves are replaced or repaired per year. Our work within the EurValve project supported the implementation and validation of a model-based decision support system (DSS) for aortic and mitral valve diseases that allows simulation, comparison and understanding of outcomes and risks of different treatment strategies. The underlying technology developed as part of EurValve has broad application to other cardiovascular diseases and is being further developed to support the coronary applications described above. These technologies include formal approaches to sensitivity analysis to account for uncertainty in clinical data acquisition and to provide confidence measures to assist clinical decision making.

Improved understanding of (patho-)physiology

Computational modelling approaches are also developed within the Medical Imaging workstream to address fundamental scientific questions around the physiology and patho-physiology of the cardiovascular system. These activities often involve the processing of imaging data in order to establish the local anatomy to accurately assess fluid dynamics effects and their interaction with the relevant biological systems. Applications include the response of endothelial cells to local variations in blood flow, particularly within stented arteries, and the role of venous valves in controlling blood flow, both in the valve region and as part of the overall contribution to venous return.

Novel computational techniques

The translation of computational modelling to support clinical decision making requires the development of novel techniques to address the demands of the clinical environment. These include the challenges of providing results from complex modelling frameworks within timescales appropriate for delivery within current clinical workflows. The Medical Imaging workstream addresses these challenges through the development and application of approaches spanning reduced-order modelling, data assimilation and Artificial Intelligence (machine learning) techniques.

People, Projects & Publications

People

Professor Julian Gunn

Professor of Interventional Cardiology
Clinical Medicine, School of Medicine and Population Health

Interest: from images to physiology.
Better decision making for treating patients with CHD.
Personalised medicine.
Bring computational modelling into mainstream cardiology.
Foster new researchers.

Email: J.Gunn@sheffield.ac.uk
Telephone: +44 114 215 9531

Interested in technology translation - clinical application.
Imaging performance quantification (EPSRC PhD).
Virtual reality (BRET).
3D printing.
Machine Learning.
Training.