Paves the way for Horizon 2020 project on Cardiac Simulations

Cardiovascular diseases are the most frequent cause of death worldwide. Cardiac arrhythmia, disorders of the heart’s electrical synchronization system, give rise to half of them. To overthrow technical limitations in simulation, biomedical engineer Axel Loewe joined forces with Hartwig Anzt and his team who specialize in algorithms for high performance computing. Together, they decided to tackle the challenge of designing algorithms and software that are capable of simulating the electrophysiology of the human heart on cellular level. Starting out with a YIN grant, the effort soon evolved into something much larger: a joint Horizon 2020 project together with 9 partners from 6 European countries.

Computer models are essential to understand the behavior of the human heart and its diseases. Like the openCARP simulation framework co-developed by Axel Loewe and his team, they are already very sophisticated and widely used. However, current models are not powerful enough to take the heart’s 2 billion (!) individual cells into account. They, therefore, assume that hundreds of cells are doing approximately the same. Due to this limitation, existing models cannot represent the events in aging and structurally diseased hearts. In both cases, reduced electrical coupling leads to large differences in behavior between neighboring cells– with possibly fatal consequences.

After getting promising early results on accelerating the simulation workflow, Axel Loewe and Hartwig Anzt teamed up with 9 partner institutions from 6 European countries and successfully applied for the MICROCARD project. From the Horizon2020 EuroHPC initiative, they received a total of 4.3 million euros in funding.

Cell-by-cell modeling of the heart poses a mathematical problem that is 10,000 times larger and much harder to solve than what has been done so far. A challenge that needs to be addressed by exascale supercomputers, which will be more
powerful and probably run on ultra-parallel computing elements like graphic processing units.

Even though the MICROCARD project is a multi-national project, the research groups of the YIN members are in the spotlight: The open- CARP software co-developed by Axel Loewe at the Institute of Biomedical Engineering will be used as central modeling framework. Moreover, the Ginkgo software ecosystem, Hartwig Anzt developed at the Steinuch Center for Computing will provide the basis for developing the powerful numerical methods that are necessary to realize the large-scale cardiac simulations.