In an effort to better understand cardiac and other illnesses, researchers from the Université de Montréal (UdeM) recently used a supercomputer to conduct what they believe is the largest mathematical simulation of the electrical activity of a human heart – a 2 billion element model.

Previously, the world’s largest simulated hearts had a few million elements at most, according to the researchers. The UdeM simulation was up to 1,000 times more detailed than previous models, they said.

The simulation was performed on a computing system from SGI (Sunnyvale, California). According to UdeM, the 768-processor SGI Altix 4700 is the largest shared-memory computing system in Canada and is used by hundreds of Canadian researchers.

To put it into perspective for those less familiar with supercomputers, Michael Brown, director of Sciences Markets at SGI, compared the power of the SGI Altix 4700 to that of a standard personal computer.

The Altix 4700 is a “pretty big system,” he told Medical Device Daily, with 768 processor cores – compared to most PCs that have only one processor core. The Altix 4700 also has 1.5 terabytes of memory, Brown noted, compared to a typical PC that has about a gigabyte of memory. Thus, he said, this system has about 750 times the computing power of a standard PC.

Mark Potse and Alain Vinet, of the UdeM’s Institute of Biomedical Engineering (Toronto, Ontario), routinely use 60 to 100 of these processors to run their simulations of the human heart. Last October, Potse and Vinet had the opportunity to use the entire SGI Altix system and its 1.2TB of shared memory to solve what they say is the largest, most detailed heart model ever.

“We have been using the model code for research and not really developing it further, but after the success of the trial I am now thinking about improving the model, making it much larger and much more detailed, and attacking other diseases that we couldn’t handle before,” Potse said. “It’s a very complicated model and it’s much, much easier to write parallel programs on a shared memory machine.”

The researchers simulated 5 milliseconds of activation in a tissue block that included some properties of a real heart, such as fiber, running in different directions. The simulation solved a system of 2 billion equations a dozen times. The test took two hours.

A full heartbeat, Potse said, would take two weeks to simulate, and his team cannot claim use of the entire supercomputer for such lengths – at least not yet.

“This was a test to see if the simulation works and to determine that, if we have a much bigger machine, our software will be able to run more efficiently,” Potse said. “This capability is really for the future when we can use this size of machine on a regular basis, but with the Altix system we have made the heart model of the future today.”

According to the researchers, discovering the electrical triggers of various kinds of heart disease could lead to earlier diagnosis and new treatment breakthroughs. In order to understand what the mechanisms of the particular disease are, the heart must be modeled with enormous detail, they said.

“Their model took up the entire system, but was still only a fraction of the size of an actual heart,” Brown said. “They think they need a system that’s about five to eight times as large as the system they have” to simulate a full heartbeat.

This means the larger, future generations of these supercomputers might someday be able to perform a reverse analysis to help doctors diagnose heart problems sooner, Brown said.

“If you can have a predictive model that says this type of pattern probably indicates this type of problem, you can do a reverse solution from the wave into the actual electrical flow through the heart, so you can diagnose something before someone has a heart attack,” he said.

The research may also help device makers develop new technology to treat a particular disease, he added.

“If you have an accurate electrical model of the heart there may be a significant opportunity for developing a new device that will increase not just the life expectancy, but also the quality of life for the patient.”

Potse and Vinet are not the first to use an SGI computing system for biomedical research.

Brown told MDD about some work being done at the University of Utah (Salt Lake City) for an electro-simulation program that would be able to simulate electro flows within the chest cavity. The researchers there developed a code for implantable cardiac defibrillators, he said, to help clinicians determine exactly where the ICD should be placed for an individual patient.

“For 25 years SGI has fueled biomedical innovations, accelerating scientific research by reducing time to insight in projects as varied as the mechanics of HIV protease, genomic correlation for cancer research, and 3D simulations of surgeries,” Brown said. “The SGI Altix system now bolsters the future of heart disease research by proving that expanded calculations of 2 billion elements are not only possible, but will become the norm.”