Magnetic Induction Tomography (MIT) is a relatively new, non-invasive imaging technique which has applications in both industrial and clinical settings. In essence, it is capable of reconstructing the electromagnetic parameters (permittivity, permeability and conductivity) of an object from measurements made on its surface. An MIT device consists of two sets of coils placed around the boundary of the object to be imaged. The first set of coils is used for the purpose of excitation, and by passing a current through each coil in turn, a primary magnetic field is created. The second set of coils is then used for measurement. This procedure causes an eddy current when each of the primary magnetic fields interacts with a conducting body inducing secondary magnetic fields, and hence voltages, that are measured in the second set of coils. Enabling MIT to take the step from being an experimental technique, which has already received some clinical interest, to become a viable imaging technique for the detection and monitoring of conditions, such as cerebral stroke, requires a step change in the quality of the reconstruction of the electromagnetic parameters and, therefore, an improvement of the computational approach used for the solution of the inverse problem. To achieve this we propose to solve the inverse Maxwell problem with a variational algorithm. Although a proof of concept of this work exists, in order to make this algorithm effective in a clinical environment, and hence applicable to the MIT problem, an implementation using high performance computing is needed, this research proposal aims to address this issue.

The primary function of the cardiovascular system is to supply blood to all parts of our body and stiffness of the tissue structures transporting the blood play a critical role in the optimal functioning of the system. The heart valves are a perfect example, which open and close over three billion times in a human life span. In spite of being robust, in our ageing society many valve-related diseases are becoming a major health problem; the stiffening of the valves leads to unwanted resistance to blood flow making it harder for the heart to pump blood at the same rate, which sometimes leads to heart failure and death. Calcific aortic valve disease (CAVD) is one such disease that progresses through accumulation of calcium within the valve tissue and affects 5% of population older than 75 years. 4 million people in the 75-84 age group are projected by 2018 and the population beyond the age of 85 is set to double by 2028. Thus, with our aging demographics, valvular diseases have been compared to an epidemic. In this study, we propose to develop a computational tool that will help identify patients at a higher risk of CAVD at an early stage of development. Based upon clinical images of a patient's heart valves and experimental results already collected by our collaborators, we will formulate a pipeline that uses the valve's movement as an input and predicts the speed and severity of its calcification. This will allow close follow-up of high-risk patients and timely intervention before the complications arise. Usually, the patients at an advanced stage of disease are recommended for valve replacement surgery; however, the patients who are seen unfit for surgery have a survival rate of merely 32% after 5 years from the disease onset. The tool developed in this project will tremendously help improve the survival rate of those patients. Furthermore, the new insight obtained from this work will help us improve the design of medical devices such as artificial heart valves and blood pumps, since currently used devices have limited durability because of valve calcification or related issues.

Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.

The aortic valve ensures uni-directional flow during the cardiac cycle, by allowing the stroke volume to be ejected from the left ventricle into the aorta during systole, and by preventing backflow from the aorta into the left ventricle in diastole. There are two main reasons for valve malfunction, 1) regurgitation (retrograde flow) and 2) stenosis (flow obstruction), which are often combined to different extents in patients. Degenerative (age-related) aortic stenosis (AS) is the most prevalent cardiovascular disease in developed countries after coronary artery disease and hypertension and is curable by open heart surgery (aortic valve replacement or, more rarely, repair).Germane to all these clinical problems is the accuracy with which the severity of AS is assessed in clinical practice. From a fluid dynamics perspective, the ideal method for quantifying AS would be to measure the energy 'loss' caused by the high-velocity flow jet across a narrow, irregular orifice and in particular by the turbulent area downstream where the jet expands. However, accurate measurement of the energy 'loss' and correlating this with clinical outcomes is fraught with difficulties. Clinicians therefore rely on two well-tested, but nevertheless imperfect, measures of AS severity: pressure gradients (PG) and effective orifice area (EOA):1) PG is a good measure for the energy loss and can be measured invasively, by passing across the aortic valve using a catheter connected to a pressure gauge or wire. The drawbacks of PG is that the procedure is invasive and that PG is flow-dependent, which requires it to be indexed when used as an assessment criteria. 2) The EOA is an alternative measure for AS severity that distinguishes between smooth and sharp constrictions. It represents the cross-sectional area of the vena contracta just downstream of the valve. The EOA is less flow dependent than PGs and considered a good measure for the energy loss caused by the stenosis. Furthermore, non-invasive fast Doppler measurements are used to determine EOA, which makes this quantity the preferred one in clinical practice. However, some major assumptions are made for the calculation of EOA, i.e. the flow jet is axisymmetric with a uniform profile and is considered flow independent. These assumptions can be questioned for the distinct three-dimensional geometry of the aortic valve, the asymmetry of many diseased valves and the incompressible turbulent flow. Hence, the aim of the proposed work is to elucidate the effect of these assumptions using computational models. Hence, this research initiative will aim to use three-dimensional heart valve models for a better assessment of stenosed aortic valves. The valve geometries will be extracted from echocardiographic data alongside the measured flow for boundary conditions. The influence of the turbulent expansion area on the jet will be evaluated using rigid opened valves or fluid-structure interaction models. The turbulence models (URANS) will be validated using echocardiographic data on the flow and pressure field. The shape, profile and direction of the flow jet through the valve are then analysed and linked to the transvalvular pressure gradients. Current clinical assessment criteria, related to the flow and pressure field, will be re-evaluated based on the modelling results.

Over the last five years the interest in developing patient-specific numerical solution to human body related problems has grown tremendously. This is due to the fact that both computing power and appropriate tools needed to carry out such studies have been emerging over the last few years. Although there are a large number of difficulties remain to be addressed, the patient-specific numerical modelling has great potential to study and understand several aspects of human body related illnesses, which are otherwise not possible. For instance, a detailed and prolong flow structure near an aortic aneurysm is only possible via a fluid dynamics study. Such a flow pattern and associated forces will help the surgeons to plan a surgery on an aortic aneurysm. Patient-specific studies will also give us the post-operative conditions a priori to a surgery and help the clinicians to make decisions. Many other examples of biomechanics, respiratory systems and urinary tract can be studied in a patient-specific sense. In short, a patient-specific study constructs a full picture from minimum available patient-specific information.The proposed network will bring a group of people from different disciplines together to address the difficulties faced by patient-specific modelling community and support a faster growth in this area. The network will pay particular attention to exploring the possibility of providing support to NHS trust hospitals. At least four formal workshops will be held during the proposed period of the network to move the research forward in the area of patient-specific modelling. A dedicated webpage will be developed and hosted from Swansea. This webpage will have a robust database for registered participants to upload and share patient-specific modelling related material. All the attempts will be made beyond the project period to sustain the network. This includes conducting larger workshops, approaching other funding agencies, charities and private medical industries.