July 30, 2018 — A team of researchers lead by Mark Potse from Inria Bordeaux Sud-Ouest is using high performance computing resources to study heart arrhythmias and help doctors in the diagnosis and treatment of patients.

HPC is enabling enormous advances in new therapies: scientists rely heavily on HPC for understanding the nature of diseases, discovering new drugs, and moving to precision medicine, customising therapies to the specific needs of a patient.

Cardiac diseases are among the biggest causes of death in Europe, and almost half of these deaths are linked to racing heart rates as a result of ventricular arrhythmias (abnormal heartbeats). At particular risk of this are both heart attack survivors and people with hereditary syndromes that can trigger fatal arrhythmias (even if they have not experienced any symptoms).

The contractions of a healthy heart are highly optimised movements that depend on an electrically activated mechanism. This mechanism is operated by around twenty different types of molecular “machine” in the external membrane of each cardiac cell. Although very effective in a healthy heart, the mechanism can trigger serious problems in the event of illness. For example, the activation can turn in a circle in a healed muscle or abnormal pulses can be triggered at a high rate by diseased cells.

Researchers from Inria and Université de Bordeaux Team within IMB (Institut de Mathématiques de Bordeaux) and IHU LIRYC (L’Institut Hospitalo-Universitaire de Rythmologie et modélisation Cardiaque de Bordeaux), used scientific computing to understand this mechanism, making use of mathematical models that were developed based on the results of experiments on cardiac cells. Already very complex in a single cell, these models require high performance computing resources to simulate the interaction of the two billion cells in a human heart.

This work took place at IHU Liryc, a research institute dedicated to cardiac arrhythmias, with one of the largest clinical teams in the world specialising in the treatment of these diseases. The modellers used nine million core hours on GENCI’s supercomputers, including Turing, Occigen and Curie, and worked closely with cardiologists as well as with biologists, physiologists and other specialists to cover all the different aspects of cardiac arrhythmias. Thanks to this collaboration and a resource allocation from PRACE, it was possible to produce work that has a real impact on the diagnosis and treatment of patients.

---

Source: European Commission