Challenges of data driven modelling in cardiac dynamics

The myocardium is an electrically excitable medium that supports various types of excitation waves, including stable or chaotic spiral waves that cause life-threatening arrhythmias such as ventricular fibrillation. Developing less invasive low-energy methods to terminate these arrhythmias requires a deeper understanding of the underlying chaotic spatiotemporal dynamics and novel control methods.  Another challenge is the limited experimental and clinical observability of cardiac electromechanical motion due to measurement limitations that require powerful data analysis tools. In this talk, we will discuss these aspects of cardiac dynamics and demonstrate how data driven modelling can provide new insights and advances in the diagnosis and treatment of cardiac arrhythmias.