On Research:
Computer Models
Can Help Cure
Cardiac Ills

The diagnosis is severe arrhythmia, abnormal electrical activity that could kill the dog within minutes. Winslow gave the dog this malady by manipulating the numbers that created its heart--a three-dimensional model that exists only inside the computer. Now, he changes a few numbers again, this time to adjust the microscopic gates or ion channels that regulate electrical excitability in cardiac cells. Moments later, the heart is beating in a slower, more regular fashion. For this computer-animated organ, "death" is no longer imminent.

Human hearts can also benefit from this high-tech marriage between biology and computer technology, says Winslow, an associate professor of biomedical engineering at Hopkins. Using a highly detailed computer model that mimics the way a heart works- -down to the subcellular level--he studies serious cardiac disorders and mathematically "tests" the drugs that might cure them.

He begins by translating the heart's physiological functions into numeric formulas, using the latest data collected by biologists. Then, by making small changes in the model, he can see how certain enzymes, proteins and other molecules make the heart beat properly--or improperly.

Using this model, Winslow is looking for medicines that could prolong the lives of millions of people suffering from congestive heart failure. His experiments have already shown that certain drugs used to control high blood pressure might also prevent sudden cardiac death. Winslow and his key research partner, Denis Noble, have formed a company, Physiome Sciences Inc., to pursue commercial applications of this software and to market drug leads discovered by the team.

Noble, a professor at England's University of Oxford, developed the first mathematical models of electrical activity in the heart more than 30 years ago. By building on Noble's work and creating even more elaborate computer models, Winslow believes the team is breaking important new ground. "Before our project," he says, "no one had ever simulated electrical arrhythmias in a three-dimensional model of the heart and then used it as a vehicle for testing drug actions."

In recent months, Winslow has discussed this research before the International Union of the Physiological Society in St. Petersburg, Russia, and at a conference on Computational Biology of the Heart in San Diego. His current model replicates a dog heart, which functions much like a human one. But Winslow says, "The methodology will also work with other organ systems and tissues."

This technique--using numerical models to study biological functions--dates back to the early 1950s, when two British scientists used crude hand-cranking calculators to come up with mathematical equations representing the electrical activity of a squid's nerve. Today, advances on two scientific fronts have made this area of research even more fruitful. First, biologists are collecting far more detailed information about how cells, and even genes, interact to determine a person's health. At the same time, computer technology is much more powerful and accessible, making it easier for researchers to compile and manipulate these complex findings.

"There's just an explosion of cellular and molecular data on the properties of heart tissue," Winslow says. "In a sense, the models have not kept pace with this explosion of data, and there's a real need to create ever more biophysically detailed models of individual heart cells that incorporate all of this information."

By using such computer models of the heart, Winslow says, pharmaceutical companies will be able to dramatically narrow their searches for life-saving medicines--and save millions of dollars now spent on conventional trial-and-error methods. "If you can tell a company to search for a drug that has a specific effect on a particular ion channel," he says, "that's important, because once these companies know what kind of drug to look for, they have the technology to screen more than 10,000 compounds a day in an effort to find such a drug."

This technique could also lead to breakthroughs involving other organs. "Models are tools for discovering the functions of biological mechanisms," Winslow says. "The approach can be used beyond just the heart. It could be applied to other diseases, other biological systems. We're using the heart as a first example, sort of a jumping-off point."

Some of Winslow's computer simulations of electrical activity in the heart can be found at the following World Wide Web address: http://www.bme.jhu.edu/~rwinslow/.