A multi-domain biophysical computational model of cardiac action potential propagation in cell monolayers

The goal of this thesis is to develop a biophysically and geometrically accurate discrete-cell computational model of cardiac action potential propagation in HL1-6 cell monolayers. HL1-6 cells are a murine atrial cell line which have slow conduction velocity in a monolayer enabling propagation to be captured experimentally at the cellular level, which is crucial in validating simulations as they shift towards advising clinical treatments. The biological model is first characterised by means of recording propagation using fluorescence microscopy. HL1-6 cell monolayers exhibit self-sustaining activity similar to re-entry/rotors observed in vivo. This activity is found to require a minimum area in which to initiate, and remains anchored around distinct lines of conduction block/slowing. A Bayesian calibration approach is investigated to capture biological variability and experimental uncertainty in action potential models. In retrospective analyses of two human atrial cell models, simpler model structures generally resulted in higher confidence in parameter estimates. The method is applied to construct an action potential model of an HL1-6 myocyte, which compares well to experimental measurements of individual ion currents, action potentials and calcium transients, and exhibits automaticity akin to the in vitro model. A multi-domain computational model of action potential propagation is developed using the spectral/hp element method and validated against theoretical and experimental results. When combined with the HL1-6 action potential model in realistic monolayer geometries, meandering self-sustaining activity could be initiated, requiring sufficient area to avoid terminating on a boundary. Increasing gap junction resistance along a thin line anchored the activity in a fashion qualitatively similar to that observed in vitro.Open Acces