Johns Hopkins Magazine -- April 2000
Johns Hopkins 
     Magazine Home

APRIL 2000


3-D modeling, say innovators like Rai Winslow, will virtually transform the way new drugs are developed.
Opening photo:
Reconstruction of a rabbit heart: Researchers used custom designed software to locate the epicardial surface (blue) and two endocardial surfaces (gold and red).
Photo courtesy Raimond Winslow
APRIL 2000
Pioneers of Promise

· · · · · · · · · · · ·
The Heart That
Numbers Built

By Joanne Cavanaugh Simpson

Waves of blue sweep across the heart, colliding with one another as cardiac cells fire in a chaotic pattern of electrical activity known as arrhythmia--one of the chief causes of heart failure in the United States.

The blue is color-coded to denote excitability, and the cells-- including their proteins, enzymes, and other molecules--are the end products of mathematical equations.The potentially deadly abnormal heartbeat won't kill this patient: It's a laptop computer.

Heart disease researchers have an astonishing new tool, a computational model of the heart, known as a virtual heart. The project is led by Raimond L. Winslow, associate professor of biomedical engineering at the School of Medicine.

What the digital heart can do, Winslow and other researchers predict, is speed up the testing of new drugs or gene therapies and analyze the behavior of specific genes, proteins, and other elements within cardiac cells.

"Our three-dimensional model simulates the geometry of the heart," says Winslow, also director of Hopkins's Center for Computational Medicine and Biology, where researchers are using similar computational modeling to study microvasculature, the auditory system, and the brain's structure.

In Winslow's model, the virtual heart actually mimics a dog's heart, which is similar to the human heart and is often used in research. "We can ask questions at every level, from the role of specific [protein-transferring] ion channels to the whole heart function."

Winslow created the 3-D virtual heart model, the first of its kind, by translating the heart's physiological functions into mathematical formulas. His work has been touted as a prototype for an emerging field of integrative modeling in biomedical engineering. In 1998, the model was selected for the Smithsonian Institution's collection on Information Technology Innovations, which picked about 40 such innovative applications from 19 countries that year.

"The Johns Hopkins University School of Medicine is using information technology to make great strides toward remarkable social achievement in medicine," according to Dr. David Allison, chairman of the National Museum of American History division where the collection is maintained.

Winslow and Denis Noble--a professor at the University of Oxford who developed mathematical models of the cardiac cell in the 1960s--have created a company, Physiome Sciences Inc., to market the virtual heart to pharmaceutical firms. Such companies can spend $500 billion bringing a drug to market via trial-and-error methods. With the model, "you could make quantitatively accurate predictions of outcomes based on data," Winslow says. "It's better than guessing." Such testing also could supplement animal testing.

On the research front, the virtual heart is providing powerful data. Last year, Hopkins cardiology researchers Eduardo Marban, Brian O'Rourke, and others used the model to test proteins that affect the balance of calcium and potassium in cardiac cells (both are needed for muscles to contract properly). Would either prove more essential in maintaining a regular heart beat?

Contrary to what researchers expected, lowered production of calcium, not potassium, seemed to lead to the irregular contractions, suggesting that drugs that restore the balance of calcium in cardiac cells eventually could be used to treat heart failure.

Research possibilities are limitless. As the Human Genome Project determines mankind's genetic makeup, for example, the virtual heart could help test the role of newly discovered genes, or set up parameters for gene therapy. And digital models could be used to simulate other biological systems, such as cancer or the immune system, or to mimic the behavior of other organs in the body--research that is already being pursued here and elsewhere.

"Modeling does not provide the proof," Winslow points out. "Experiments and clinical studies do that. A model can't predict everything in a system as complex as the heart. But the model can point people in the right direction."