Mapping the Heart

Date: May 31, 2013

Natalia Trayanova: “We are trying to create a model of the heart that would work like Google Maps.”

Natalia Trayanova studies hearts that don’t work properly. She’s not a cardiologist, but a professor of biomedical engineering and member of Johns Hopkins’ Institute for Computational Medicine. Though Trayanova works closely with cardiologists, the hearts she studies are on a computer screen. 

“We are trying to create a model of the heart that would work like Google Maps,” she explains. “The goal is to be able to zoom in and out, going all the way down to the molecular (street) view or all the way out to the (global) view of the whole organ.”

To accomplish that, Trayanova must study the electrical and mechanical functions of the heart as well as the way in which cardiology is practiced. “If we can figure out which molecular changes create the big problems—and why—we may be able to better diagnose patients early on,” she says, “based on what we know about their genes and proteins.” Trayanova is working to equip doctors with the tools they need to tailor treatments to each person.

“Our goal is to learn as much as possible through these noninvasive tests that we are developing,” she says. “The more we know about heart function at both the theoretical level and the patient-specific level, the more we can improve the current therapies for patients suffering from heart disease.”

Imagine that a patient comes to see a cardiologist because of a quickly beating heart. The patient receives an MRI or a CT scan to assess the problem. Trayanova and her team then use detailed information from the scan to construct a 3-D computer model of that patient’s heart. The software they develop models treatment options to predict which will give the best outcome.

Getting to the Heart of the Problem

The heart has two major functions: electrical and mechanical. Its mechanical, pumping function is triggered by electrical pulses sent throughout its muscle cells so their contractions are exquisitely timed and coordinated. 

In a heart in which pumping has gone awry, cardiac electrophysiologists can’t tell which cells are off-kilter just by looking, so they use an electrical probe to test clusters of cells over the whole surface of the heart. Locating the problematic cells can take four to eight hours.

Replace that scenario with one in which the patient receives a noninvasive MRI scan. The scar tissue—the cells creating the problem—shows up as a bright spot on the scan, allowing doctors to navigate a hot probe to the location shown in the computer model and burn the problematic cells. 

“We applied our model retrospectively to MRI scans from patients who underwent the standard probing procedure, and its predictions matched what the cardiac electrophysiologists found,” says Trayanova. “So we are now planning clinical trials.”

—Catherine Kolf

For more, click here.