Straight from the lab: technology’s first draft.
Pollsters, public health officials, and marketing executives are among the many who use cartograms, special maps that provide a visual sense of geographic information by distorting it in proportion to a key variable. Researchers at AT&T labs in Florham Park, NJ, have created software that takes just minutes to calculate the distortion for thousands of regions (say, for all 3,066 U.S. counties), while old methods could take hours and could handle only tens of regions at a time. The AT&T technology reduces the computational job by imposing lines of various lengths and orientations across the map and performing size adjustments to the space on either side of each line. The process reconfigures geographic regions into recognizable but meaningfully transformed shapes without having to recalculate the whole map. Earlier cartogram programs used different approaches: Some treated the regions like balloons that inflated or deflated in a process that worked quickly but didn’t preserve recognizable shapes. Other software that treated the data as a series of mathematical equations produced more familiar and accurate shapes, but consumed hours of processing time. Daniel Keim, the new software’s lead developer, says AT&T began using it this year and researchers are refining it while the firm pursues licensing agreements.
Name That Tune
Forget the name of a song you heard on the radio this morning? Soon, thanks to an audio search engine developed by researchers at Philips Electronics in Eindhoven, the Netherlands, you will be able to search a database and find the song simply by humming its melody. A specialized device embedded in either a desktop computer or a compact disc player will capture your voice, and algorithms will identify the duration and pitch of each note in the song’s melody, creating an audio fingerprint. The fingerprint will then be fed into a search engine that uses pattern recognition software to find a matching song stored on a compact disc, computer hard drive, or even the Internet.
Other audio-search engines in development might not work well for people who are even slightly tone deaf, but Philips’s technology works for people with musical training and “people who can’t sing, like myself,” says Boris de Ruyter, a senior scientist at Philips. The company hopes to market the technology in consumer products within two years.
Feel-Good Drug Making
A system that lets researchers touch, feel, and prod 3-D molecular images could improve scientists’ sense of how molecules interact, aiding the development of new drugs. Texas A&M University biochemist Edgar Meyer and his colleagues Stan Swanson and Jennifer Novak are developing simulation software that will make just such a system possible. Their software calculates millisecond-by-millisecond changes in the behavior of the interacting molecules; a commercially available joystick-like interface translates that 3-D information into a tactile sense of pushing, pulling, or resistance. Because many therapeutic agents work by binding tightly with disease-causing molecules, the software could be a critical tool that allows researchers to virtually check-and modify-the fit between molecules. Meyer plans to partner with one or more pharmaceutical companies within the year.
DuPont researchers have developed a flexible textile that conducts electricity. The advance could facilitate the seamless incorporation of electronic devices into clothing. Each fiber of the new textile consists of a core of DuPont’s ultrastrong polymer Kevlar covered with a layer of electrically conductive material such as silver or nickel. Bundles of these fibers are coated with a second polymer for protection during washing. Although many textiles for wearable electronic devices are under development, the DuPont material is the first that combines high electrical conductivity with flexibility sufficient for weaving and embroidery. DuPont’s fiber is now commercially available and was recently used by a Finnish sportswear company to make a prototype cold-weather survival suit-complete with an embedded heart monitor, a body-temperature sensor, and a Global Positioning System device.
Atlanta-based Altea Development is creating what has long been a diabetic’s fantasy: a system that delivers insulin through skin patches rather than by injection. First the patient places on his or her skin a handheld device bearing an array of tiny filaments. Each of the filaments delivers a three-millisecond burst of heat that vaporizes dead and dying cells and creates a micropore in the skin’s outer layer, which normally blocks absorption of large molecules such as insulin. Because the pulse of heat is too brief to affect the underlying living cells, the process is painless. The patient can then replace the device with an insulin-filled patch; for the next 24 hours, the insulin diffuses into the skin and is absorbed into the bloodstream. Alan Smith, Altea’s vice president of research, says that preliminary tests with human subjects have been successful, and he expects more extensive clinical trials to begin by the end of this year.
Amit Lal is looking for power-in doses small enough to run microscopic machines. The Cornell University electrical engineer and his colleagues have found a way to use nuclear energy to power such micro-electromechanical systems as those used for sensing or communications. Making batteries small enough to work with microscopic machines has been a challenge, but a speck of radioactive material, says Lal, could power such a device for more than 75 years. Lal’s system comprises a tiny copper cantilever that hovers above a thin film of radioactive nickel. The nickel emits electrons that hit the tip of the cantilever, causing it to deflect and swing back. This cyclical motion could be used to move a second mechanical component, or to generate electrical current. Using that current, nuclear-powered sensors embedded in the walls of buildings could run autonomously for decades, says Lal. He and his Cornell colleagues have started a company called Lifesonics in Ithaca, NY, to commercialize the technology. Lal hopes to have a prototype nuclear-powered temperature-and-pressure sensor ready within two years.
The Pulse of Cable
Ultrawideband technology, which uses timed pulses to transmit data, is rapidly emerging as a breakthrough in wireless: it provides the possibility of sending more data faster than any alternative. San Diego startup Pulse-Link wants to make ultrawideband the hot new technology for cable television as well. Requiring no changes to existing cable-television infrastructure, Pulse-Link’s system calls for as little as one new piece of equipment at the cable company’s facility and an additional user-installed box in each subscriber’s home. Pulse-Link’s equipment can encode and decode data that are sent in approximately nanosecond-long pulses and spread across all the radio frequencies used by standard cable-television. This technology, which won’t interfere with conventional signals, should double the amount of information that can be sent over the cables and allow cable operators to deploy bandwidth-hogging high-definition television signals without giving up existing channels.
Today’s hefty automotive cooling systems, which can barely keep up with the heat produced by our car engines, may be inadequate for cooling the more demanding hybrid and fuel-cell vehicles of tomorrow. For a boost in cooling power, Stephen Choi and his team at Argonne National Laboratory have added nanometer-size copper particles and carbon nanotubes to radiator fluids such as ethylene glycol. Because the solid particles conduct heat a thousand times better than most liquids, researchers have been dreaming of such liquid-solid mixtures for decades. But it is only recently that they’ve been able to create particles small enough that they don’t settle out of the fluid or abrade engine blocks. So far these nanofluids have demonstrated more than double the cooling capacity of typical coolants; that could translate into smaller, lighter cooling systems and better gas mileage. Argonne is partnering with several companies to develop a cheap means of producing the nanoparticles and hopes to bring nanofluids to market within five years.