Tung is working on models of cardiac electrical activity that could lead to a better understanding of, and treatments for, heart rhythm maladies, which afflict millions of Americans and often occur when the heart's electrical signals don't trigger the right result. Lately, he and colleagues in Baltimore have been working with heart cells from fetal rats, trying to map and then simulate on computers the way electric waves propagate through the tissue.

"It's arrhythmia in a dish," said Tung.

Cardiac cells belong to the same family as nerves, most kinds of muscle and other "excitable" tissue. Their electrical activity involves the flow of various ions, including calcium (which triggers contractions, too) sodium and potassium, across the cell membrane and through gated protein channels.

"The heart controls the activation and synchronized activity of the cells to produce efficient and coordinated contractions," said Tung, who came to Johns Hopkins in 1986. "In an arrhythmia, the coordination is lost or it's inefficient, and the efficiency of the pump goes way down."

This particular pump, as most high school biology students know, consists of four chambers, left and right ventricles and left and right atria. Contractions — in lay terms, "beats" — originate with a pulse from the sinus node, the heart's own pacemaker, located in the right atrium. This jolt triggers a cascade of electrical signals that cross into the left atrium, where they get picked up by the atrio-ventricular, or A-V, node, which in turn passes them along to muscle fibers in the ventricles.

Deviations from this pattern, or variations in its timing, can lead to roughly a dozen forms of arrhythmia, from ventricular tachycardia, in which the heart speeds up, to atrial fibrillation, erratic impulses from the atria to the ventricles that produce an irregular beat. Symptoms of arrhythmia typically include fatigue, dizziness and chest pain, but some people with the condition feel no signs of it until it's too late. Arrhythmia significantly ups their risk of stroke, heart attack and sudden cardiac death.

Tung believes work like his might one day improve not only detection but treatment of arrhythmia. To do so, one must know one's enemy.

"We're trying now to build a computational model that is a one-for-one correspondence" between real cells in a dish and their representations on a computer monitor, Tung said. "As far as I know this kind of rigor hasn't been attempted before."

Tung's group is cultivating monolayers of cells from rat ventricles that allow them to observe the way the tissue handles electricity in two dimensions. Using a fluorescent dye to stain the tissue, and a 64-fiber optical array, they can monitor the electrical potential of the cells as they fire.

"We're able to obtain maps of the electrical properties of cells during normal and abnormal activity," Tung said.

Alok Sathaye, 21, of Warren, N.J., one of Tung's undergraduate students, recently completed the first phase of a computer program to simulate the rat cells. Sathaye's code incorporates what's known as a FitzHugh-Nagumo model, a system of mathematical equations that describe the active and recovery states of heart cells.

"The major reason why I am interested in Dr. Tung's work is that the cardiac system is one of the few systems within the body which can be represented by a wide variety of engineering modalities," said Sathaye. "It is a mechanical and electrical system. The work that Dr. Tung is doing is a good combination of computer modeling, instrumentation design and cell biology."

Tung said that Sathaye is striving "to build a model using these cells as a first step toward making a close parallel" for living tissue, Tung said. "The next step, which is nearly complete, is to replace the FitzHugh-Nagumo equations with cardiac equations that are similar, but much closer" to reality.

"By making a closely parallel model we hope to be able to validate the model and use it in computational experiments," Tung said, adding that computer models are "much easier to do and much more powerful" then animal experiments. A good computer simulation can manipulate many more parameters than could be controlled or studied in a live-animal model, not to mention a human trial, which would be both ethically and practically unfeasible.

Tung termed the work "a way station toward trying to understand the 3-D behavior" of heart muscle; "3-D is really cutting edge, verging into the area of tissue engineering. You have to worry about the substrate, the scaffold on which the cells are grown."

Tung has been working in the field of cardiac electrophysiology since his graduate student days in EE at the Massachusetts Institute of Technology. Indeed, Brad Roth, a biomedical engineer at Oakland University in Rochester, Mich., said Tung's doctoral thesis is "probably one of the most widely cited theses" in the field.

Roth is also studying how to model the electrical properties of heart muscle, using what's known as the bidomain model, a set of equations that allows researchers to factor in the tissue's anisotropic nature. Heart muscle, it seems, likes to play games with polarity. Stick an electrode into a tissue sample and give it some juice, and the electricity will propagate in different directions and in different amounts, Roth said.

"That can lead to some interesting features. A single electrode stimulation gives depolarization under the electrode, but then you get hyperpolarity nearby on either side.

"Our goal is to improve pacemakers and especially defibrillators," said Roth, who noted that the implantable beat keepers are far simpler than the shock devices, which is essentially what defibrillators are.

"Even though companies have made these defibrillators and they're selling thousands of them, people don't really understand the basic mechanics of them," Roth said. "The real questions are how the applied electric fields polarize the cardiac cells."