Some efforts entail innovations in device-level manufacturing, packaging and materials. They promise to stretch the boundaries of communications-especially RF and antenna design-as well as distributed computing. But years of work are ahead before today's promising lab projects and field trials can become a reality.

Perhaps the most tangible advancements today in what loosely may be called wearable computing are in the field of RF identification to track products through the supply chain. This fall, the Uniform Code Council that maintains bar code standards is expected to publish version 1.0 of a new RFID standard developed by the Auto-ID Center at MIT that could help push the technology toward use in billions of products.

"This industry is going to ruffle some feathers," said Sanjay Sarma, the research director of the MIT center that developed the technology behind the pending standard and will now move on to R&D work in the field. "It's a wonderfully rich area," he said.

"Eventually, you have to wire up shelves with antennas. If we are going to do that, we may have to print antennas on laminated paper. The economies of scale that have driven RF technology will change," Sarma predicted.

The chip packaging industry in general "has morphed around high-value ICs. In the future, packaging must become more like the printing industry," to enable RFID, Sarma added.

Such underlying technical advances could also breathe new life into the maturing computer industry. "IT has never affected store clerks. Operations have always eluded the net of information. The outlook is unpredictable because I think we will be surprised by the new apps that come up," he said.

Texas Instruments, which claims it has sold nearly 250 million RFID tags in the company's 15 years in that business, has not taken an active part in the MIT center's work. Nevertheless, Bill Allen, a marketing communications manager for the TI group, welcomed the effort at standardization.

"Large customers like Wal-Mart don't want to adopt proprietary technology. I don't think any major retailer is going to move forward until these standards issues are settled," said Allen.

For its part, TI is taking part in a field trial in Phoenix with American Express under which TI's 13.56-MHz RFID tags are used to make payments at 175 shops by simply waving a key fob in front of a reader. "We think this payment space represents a growing market," Allen said.

"There's a bunch of companies investing significant money in developing the underlying hardware needed here," said Paolo Bonato, director of the motion analysis lab at Harvard Medical School's Spaulding Rehabilitation Hospital. Bonato recently edited a collection of papers on wearable medical devices in the May/June 2003 issue of IEEE Engineering in Medicine and Biology.

At the commercial level, a handful of startups are preparing shirts, wristwatches, armbands and rings with embedded sensors to track a variety of heart, lung, temperature and motion data for people suffering various ailments, as well as for athletes and the military.

Most devices require wired links to PDAs or PCs with wireless models in the labs. Their challenges are to integrate more complex sensors and low-cost wireless communications.

"This area is moving from a clinical lab setting to the field with a range of applications. It is still difficult to say what will be the growth in [wearable] health care [devices], but in research, the growth will be exponential for the next five years," Bonato added.

VivoMetrics Inc. (Ventura, Calif.) is one of a handful of startups in this field. The company's LifeShirt, essentially a sensor-rich T-shirt, currently is aimed at tracking breathing for ailments like sleep apnea. The company's beachhead market is using its products to manage data collection currently for five pharmaceutical trials. Andrew Behar, founder and chief operating officer of VivoMetrics, sees that as a big growth market, given that as many as 500 clinical trials are in progress at any time.

In Italy, doctors are using the LifeShirt as a tool that lets them monitor stroke patients from home, reducing the length of hospital stays. "This could be used broadly in health care in, say, five to 10 years to get people out of hospitals and back at home earlier," Behar said.

But a route to getting the LifeShirt into the hands of the average consumer is still beyond the horizon. "Doctors aren't quite sure what to do with all that data yet," he said.

The company's top priority is working more complex, less invasive sensors into the LifeShirt for measuring heart rate, brainwaves and other critical factors. Vivo is working with one company to develop a lightweight sensor that can track heart rate through clothes.

BodyMedia (Pittsburgh) aims to enable such products with its SenseWear transceiver that Roche Diagnostics will sell starting this summer in a $300 to $400 armband for weight-reduction monitoring. Chief executive Astro Teller hopes to sign on one or two more transceiver users every year to push the startup to profitability. Medtronic and Audi are already using the device in some of their trials.

The programmable SenseWear Transceiver can be integrated into most products to enable wireless two-way communications.

The Zigbee Alliance is responding to the call for low-cost wireless links on wearable medical devices. The group recently formed a medical task group that will publish in 2004 a profile for medical devices of the low-cost, low-bandwidth wireless technology.

Beyond the wearable medical devices, researchers are exploring implantable chip-level devices on many fronts. Perhaps the most challenging and fascinating of these involves neural prosthetics, implantable silicon neurons that could someday carry out the functions of a part of the brain that's been damaged by stroke, epilepsy or Alzheimer's disease.

That work is mostly in an animal-research phase now. Work in humans could be five or 10 years away, according to researchers.

One of the thorniest problems is defining an interconnect between body tissue and silicon.

"We are working with chemists and materials scientists to figure out how to coat interface devices with biological or biological-like material that will attract neurons. It could actually be sticky for neurons. It's a hard problem, and we've only started working on it recently," said Theodore Berger, director of the center for neural engineering at the University of Southern California.

Sarma: Packaging must become more like printing industry.

The Army is working on multiple antenna designs spanning various frequencies. The power/data bus would have capabilities similar to the multi-megabit FireWire and USB interfaces, linking lightweight computers, batteries, heads-up displays and sensors carried or worn on the soldier's fatigues.

"There are several fabric-based conductors we are now looking at. We think of it as something that could be transferred to commercial wearable systems," said James Fairneny, an electrical engineer, in the U.S. Army Research and Development Center outside Boston.