The main scientific results achieved by the network are clustered in 3 research areas. ESR effort focussed on numerical and experimental techniques, including testing and validation, applying them to address clinical and industrial needs not met by current systems.
Cardiac tissue function and cardiac support
This combined numerical methods with ultrasound imaging and use of the ex vivo PhysioHeart platform. Analysis of experimental ultrasound data to retrieve strain measures from dynamic cardiac images was combined with in silico assessment of the same technique using a known strain field to characterise error propagation in strain measurement. Physiological processes in cardiac tissue were examined using the PhysioHeart platform (LifeTec Group). Development of 3D models of cardiac function included a structural model of the left ventricle with passive and active properties of cardiac tissue. A database of 200 synthetic pathological finite-element heart models from 5 real healthy left ventricle geometries assessed sensitivity of specific strain-based parameters in locating myocardial infarcts. Detailed modelling of cardiac electrophysiology used a novel experimentally-calibrated population of models methodology to detect long term changes in the electrical activity of the heart supported by experimental Physioheart data to inform virtual populations of models.
Cardiovascular haemodynamics - pathology and intervention
This research applies numerical methods to analyse flows in the heart, aorta and medical devices that support heart function. A computationally efficient moving-mesh method (ANSYS CFX) to account for wall motion in CFD simulations of aortic dissection was developed. Further work deployed modelling tools in combination with patient-specific data. Development of novel approaches to analyse the performance of blood oxygenators used simulation with explicit meshing/porous media to represent the fibre bundles and simulation of single phase/multi-phase properties of blood, reporting implications for future oxygenator design. Methods for improved segmentation to obtain aortic geometry from medical images were extended to inform model boundary conditions. Novel Reduced Order Modelling techniques were shown to give good agreement with full 3D CFD simulations for idealised flow conditions.
Image-based diagnosis and imaging quality assurance
This research had an imaging focus, covering cardiac ultrasound, angiography and flow phantoms. Experimental and theoretical work towards a ring vortex phantom investigated stable and reproducible experimental vortex ring generation in both air and water, exploring implications for phantom design. A semi-automatic method for the reconstruction of 3D stented coronary artery models exploited OCT and 3D printing, the high fidelity 3D geometry of the stented phantom was successfully used for CFD simulations with ANSYS Fluent software. Gabor filters were developed to improve quantitative processing of TO-US images in both axial and lateral directions, increasing the robustness of the Phase Based Motion Estimator method.
In conclusion, VPH-CaSE aimed to realise the full potential of the ESRs. This included opportunities to experience academic, industrial and clinical environments through exchanges and provide new career perspectives, managed through a Personal Career Development Plan that documented collaborative meetings and longer term secondments. Wider issues were also recognised, acknowledging the values of integrity, transparency and ethics in research. VPH-CaSE has exposed research outcomes through formal scientific publication, a project website, newsletters and video content. Numerous public engagement and outreach events were also delivered. Collaborative links developed within the project period are being sustained beyond the project period through ongoing publication of research outcomes.
