From pixels to particles: multi-physics meshless model of the heart

The functioning of the heart, which main role is pumping blood to the rest of the body, is the result of a complex combination of multi-scale physical processes from the molecular to the tissue level. The predominance of animal experiments to understand the cardiac function has a high cost and associated ethical burden. There is a growing interest for non-invasive ways of extracting measurements. Owing to improvements of numerical methods and computational infrastructure, cardiac models might provide complementary clinically-relevant information in a non-invasive way. The majority of cardiac models use mesh-based numerical techniques, whose performance depends on the construction of a well-defined mesh that represents the high complexity of anatomical structures such as the heart. This thesis presents a derivation of a fully coupled multi-physics meshless model of the heart as an alternative and provides robust evidence of its ability to simulate the heart's behavior as observed in experimental measurements.