Magnetic measurements are also used to study the brain and the heart. Nerve activity in the brain generates very weak magnetic fields--about 10 orders of magnitude smaller than the earth's. Measuring this weak biomagnetism requires highly sensitive magnetic detectors called SQUIDs, which in turn require superconducting materials. The most sensitive SQUIDs must be cooled to within a few degrees of absolute zero with liquid helium; they cost about $2 million.

Kitching's magnetometers are nearly as sensitive as SQUIDs and can operate at room temperature. He says that they are currently sensitive enough to measure magnetic fields from the heart but not from the brain. "Fetal heart monitoring is getting a lot of attention in the medical field" but is difficult because it's not possible to place electrodes directly on a fetus in utero, says Kitching. "Electrical fields don't get to the surface unaffected [by the mother's tissues], but magnetic fields do," he says.

David Cohen, who made some of the first measurements of biomagnetism in the 1960s, says that Kitching's magnetometers "may get to the point where you can measure the heart," but he is skeptical that they will be used to study brain activity. He doubts that a device using the NIST sensors to detect biomagnetism would end up being any cheaper than those that rely on SQUID.

Another potential use for the sensors is in future MRI scanners. "For noninvasive biological measures, this could be a really interesting thing," says Yael Maguire, who, before founding ThingMagic, in Cambridge, MA, worked on miniaturizing nuclear magnetic resonance detectors, a technology similar to MRI. MRI currently requires its own room, specialized technicians, and a large, strong magnet. "The cost of access to the machines" is a problem with MRI, says Maguire. (See "Better Pictures of Proteins.") Highly sensitive, cheap magnetometers like Kitching's could be incorporated into future MRI scanners, enabling them to use smaller magnets, bringing their cost down, and potentially making them portable.

But such clinical applications are many years away. Right now, Kitching says that he's studying the trade-off between the size and sensitivity of the magnetometers and is also designing chips to carry them.