If this all sounds a bit theoretical, that's because it is. Katz and Wang expect it will be four years before their newly funded project reaches completion. At this stage, Katz can't even say for certain which injuries their system might be able to recognize, or exactly how it could treat them. Right now, he says, they're simply designing a logic gate that can distinguish between different injures--what the biomarker combinations look like and the enzymatic code to interpret them. Next they'll decide which bodily fluids would work best, and from there they can begin their electrode design.

Of the hundreds of sensors in Wang's office, he points to a few he believes might be useful models. One, meant to be rolled up into a tight cylinder, is so minute that it could fit into a tear duct. Another, larger one could have a small subcutaneous sensor that sits just under the skin. "We want something that would be minimally invasive, or, more desirably, noninvasive, that could sample tears, saliva, or sweat," he says.

The researchers have a big task ahead of them. "I think an important challenge is finding out of the things they can sense, how reliable [they] will be in a battlefield situation," says Martin Bazant, a professor of mechanical engineering at Stanford University. "Will you be able to add value to the soldier without adding weight or risk of malfunction?"

Bazant is familiar with the difficulties of designing for soldiers in combat--he was one of the founding members of MIT's Institute for Soldier Nanotechnologies--and he notes that development of the sensor itself would be a huge boon. "Having the ability to detect accurate levels of those chemicals in real time in the battlefield, reliably--that's already interesting," he says. "A medic could read it off, and use it to determine how critical a patient is, whether treatment is necessary, whether a patient should be moved to another location." However, Bazant is skeptical about the use of a fully automated system for injury sensing and drug dosing in the absence of a medic.

If Wang and Katz are successful, their project will have applications not just during wartime but in everyday medicine. Doctors are always in need of sensors that provide a more accurate picture of what's going on in a patient's body. It could be adapted to detect cardiac markers--for example, to rapidly diagnose a heart attack or a stroke. "This can be useful whenever we have something urgent that needs quick action," Wang says.