Computer predicts heart attacks before they happen
Watch how blood flows through your heart. No surgeries required.
The same algorithmic predictions that allow Netflix to know what movies you want to watch is helping researchers mash up fluid dynamics, computer science, physics, and cardiac medicine into a computer custom-built from graphics processors. Popular Science explains.
It’s called multiscale hemodynamics, and the technology lets doctors see exactly how blood moves through the heart, predicting where future arterial blockages are likely to form.
Basically, it predicts heart attacks before they happen, and it does all of this using nothing more than a simple CT scan.
Heart attacks occur when plaque builds up in arteries around the heart, leading to blockages that restrict blood flow. Plaque develops in areas where stress on the arterial walls is low – like the places in the stream where water slows and pools.
In the US, about 1.2 million people suffer heart attacks, and doctors have no idea how to predict where blockages will be. If they suspect something, the current standard course of action involves threading a camera-equipped catheter through the circulatory system and into the heart. In other words, patients undergo a surgery before doctors are even sure there is a problem.
Computed tomography (CT) scans, on the other hand, are cheap and noninvasive.
- Harvard’s Efthimios Kaxiras and colleagues began modeling the way blood flows through arterial trees from CT scans of an individual patient's heart. This allows them to look in the heart without actually venturing inside.
- To model blood, the team tapped into the resources of some of the country’s biggest, baddest supercomputers – IBM’s Blue Gene ran an early version of their models – generating fluid dynamics models of the heart.
- To shrink the software down to the clinical level, they turned to new technology that NVIDIA provides through its graphics processing units (GPUs).
- Then they programmed a GPU-driven platform capable of running the multiscale hemodynamics models at a fraction of the cost and complexity of a supercomputer.
Results that used to take weeks or months, now take only an afternoon. With every incremental improvement in the technology – in the quality of the CT scans, in the efficiency of the software, in the power of the GPUs, in the strength of the equations – the process grows a little faster.
“In five to ten years this technology is going to be so good and so well-oiled that you are going to be able to push a single button on the CT scanner and get all of this information in a few hours,” says Frank Rybicki of Brigham and Women’s Hospital in Boston. “Or maybe in a few minutes. All of the pieces are there.”
Kaxiras, Rybicki, and their collaborators are working on a clinical prototype that puts all of the computational hardware in one package so it can be easily deployed in any radiology suite.
From Popular Science.
Image: NVIDIA
This post was originally published on Smartplanet.com