Integration of different cardiac electrophysiological models into a single simulation pipeline

Clinical translation of computational models of the heart has been hampered by the absence of complete and rigorous technical and clinical validation, as well as benchmarking of the developed tools. To address this issue, a dataset containing the cardiac anatomy and fibre orientations from magnetic resonance images (MRI), as well as epicardial transmembrane potentials from optical mapping acquired on ex-vivo porcine hearts, have previously been made available to the community. Image processing techniques were developed to integrate MRI images with electrical information. Different models were tested and compared with the integrated data ¹, including: i) a new methodology to customize and regularize heart shape and myocardial fibre orientation to predict activation waves in these personalized meshes with generic conduction parameters; ii) a statistical model-constrained framework to produce maximum a posteriori (MAP) estimation of the 3D distribution of transmembrane potential; iii) a personalized simplified reaction-diffusion 3D electrophysiological model; iv) a personalized fast conduction Purkinje system with a simple Eikonal-based electrophysiological model.