It’s not quite Mary Shelley’s image of a corpse brought to life by electricity, but biomedical engineers have found a way of using electricity to bring artificial bone to life. The method could one day yield bone replacement parts.

Bioengineer Rena Bizios at Rensselaer Polytechnic Institute uses carbon nanotubes-tubular molecules that are good electrical conductors-to deliver electricity to bone-forming rat cells deposited on a piece of polymer. Researchers have long known that electrical stimulation enhances bone growth, but it’s hard to deliver the electricity uniformly: new bone tends to clump around the electrodes delivering the charge. Bizios’s technique could solve that problem, though, since the nanotubes are embedded throughout the polymer. When the researchers turned on the electricity, the bone cells grew and began to deposit the proteins and calcium that give bone its strength. That the technique worked so well “was a great surprise,” says Bizios.

Researchers don’t know yet if the approach will ultimately yield uniform bone tissue, but the results are “very exciting and very promising,” says Antonios Mikos, a biomedical engineer at Rice University. While doctors can treat small bone injuries by surgically implanting patchlike materials, they can’t yet generate the large sections of bone that would be needed to replace a hip ravaged by osteoporosis, for example. Bizios’s material, on the other hand, opens up the possibility of quickly growing large sections of artificial bone in the lab using a patient’s own cells and nanotube-wired polymer scaffolding. Surgeons could then replace any damaged or diseased parts of a patient’s skeleton with the new bone.

Gazing at an electrical meter, Yi Cui, a graduate student in the Harvard University lab of chemist Charles Lieber, waits for evidence of a remarkable feat in simple, ultrasensitive diagnostics. His target is prostate cancer. His new tool is a microchip bearing 10 silicon wires, each just 10 nanometers (billionths of a meter) wide. These nanowires have been slathered with biological molecules with an affinity for PSA, a protein all too familiar to men of a certain age as the telltale sign of prostate cancer. If the experiment works according to plan, when the PSA molecules bind to the nanowires, there will be a detectable electrical signal.

Cui washes a solution containing prostate cancer proteins over the chip. Immediately, the meter registers subtle changes, indicating not only that the device has detected the protein, but that it detected perhaps as few as three or four molecules, instantly and with minimal sample preparation-a previously unheard-of feat. The implications for diagnostics are enormous. A successful prostate cancer test must distinguish between normal and elevated protein levels. Ultrasensitive sensors like Lieber’s could discern the slightest increase; what’s more, they could do so in cheap, disposable tests that patients could use at home between visits to the doctor. “If I were at risk for a particular cancer, I wouldn’t want to take a chance and wait for some cancer cells to grow wildly out of control over a year because the previous test missed it,” says Lieber.

Though this nanowire device is just an experimental prototype, it is at the forefront of a growing effort at labs around the world to marry nanoelectronics and biology into a new field called nanobiotechnology. This hybrid discipline is producing a variety of tools-from arrays of tiny sensors that can detect specific biological molecules to microscopic systems carved out of silicon that can read individual strands of DNA-capable of providing a new window on biological molecules.

The implications for medicine and biotechnology are myriad. Besides sniffing out the barest whiffs of disease-or perhaps detecting a single spore of anthrax-these devices could provide far faster and easier diagnosis of complex diseases. For example, they could provide early warnings about heart attacks, whose calling cards are subtle changes in the mix of dozens of proteins. Alternatively, a single microchip could provide a comprehensive diagnosis from a drop of blood. And for drug researchers, nanobiotech gadgets could mean new tools for discovering and evaluating potential drugs more rapidly, by screening millions of different drug candidates at once. Some of these more ambitious goals will likely take years to achieve, but nanobiotech could lead to real devices that will begin replacing cumbersome lab-based procedures with cheap, accurate microchips in as little as two years.

These first products-chips rigged to detect a specific disease or cluster of genetic disorders-are already being developed at nearly a dozen nanobiotech startups (see ” Sensing Success “) . Larry Bock, CEO of Palo Alto, CA-based startup Nanosys [ TR board member Robert Metcalfe is a Nanosys cofounder and director. Ed.], which has licensed Lieber’s technology, predicts his company will market a commercial sensor within three years, first for use as a research aid to rapidly screen potential drugs, and later as a cheap, disposable at-home test for prostate cancer and perhaps other cancers. “People talk about all the wonders of nanotechnology but then say it’s not going to happen for another 20 years,” says Chad Mirkin, a chemist and director of the Institute for Nanotechnology at Northwestern University. “But that’s absolutely incorrect for things like diagnostics. You’re going to see products on the market in the next two years.”

Power in Numbers

Biology and electronics have long existed in separate universes. But because biological molecules, like DNA and proteins, are roughly a few nanometers in size, and because physicists and chemists are now learning how to make electronic devices on exactly that size scale, these universes are colliding. The result is a new class of devices that combine the ability of biological molecules to selectively bind with other molecules with the ability of nanoelectronics to instantly detect the slight electrical changes caused by such binding. “What’s really interesting about this technology is that it allows one to take the inorganic components that normally would be nestled inside an electrical chip and combine them with biological molecules,” says Paul Alivisatos, scientific cofounder of Nanosys and a chemist at the University of California, Berkeley.

Indeed, nanoelectronic devices like the one built in Lieber’s lab (see ” Sensitive Wire “) could do away with the elaborate apparatus now needed for ultrasensitive detection. “If you wanted to do single-molecule detection in a lab today, you would need a laser the length of a desk and a lot of sophisticated optics, chemical labels to amplify the signal enough to be able to see it,” Bock says.

Shrinking down such ultrasensitive devices enough that they could be put on chips could have numerous applications in diagnostics. Stanford University chemist Hongjie Dai, for example, has built a device that can detect glucose with a single carbon nanotube, a large carbon molecule with excellent electrical properties (see ” The Nanotube Computer ,” TR March 2002) . The glucose molecules react with molecules on the surface of the nanotube, creating electrical signals that correspond to glucose concentrations, he says. Though only a proof of concept today, such a device could be developed into an implantable glucose sensor for diabetics. In December, Dai launched Molecular Nanosystems in Palo Alto, CA, to commercialize nanotube-based devices including biosensors.

For many applications, though, what’s really needed is not a lone nano detector but a dense array of them. That way, you can rapidly look for thousands, even millions, of different biological molecules in a single drop of blood or other body fluid, allowing the diagnosis of diseases that have complex molecular signatures. One such disease is rheumatoid arthritis-an autoimmune disease with many variants, each marked by subtle differences in groups of proteins. Ideally, each variant would be fought with a slightly different treatment; in practice, sufferers today are generally treated in the same way. But, says Dai, a nano array could serve as a highly precise and discriminating diagnostic device, providing a road map for custom treatment.

These arrays of nano detectors promise advantages over existing technologies, like DNA chips, and ones under development, like protein chips. All such chips require fluorescent labeling of molecules and optical microscopes to detect the glow given off when binding occurs (see ” DNA Chips Target Cancer ,” TR July/ August 2001) . What’s more, roughly a thousand molecules must bind to each sensing element to create the glow. With nanoelectronics, no bulky, expensive equipment is needed, and instant detection of just a few molecules is possible.

Sensitive Wire

To detect a disease-related protein in a blood sample, a silicon wire just 10 nanometers wide is coated with biomolecules that bind only to that protein (below). When the disease protein binds to a molecule on the wire (inset), the wire’s conductance changes, providing an instant electric signal.

Sticky DNA

But sensors with nanoscale features can only succeed if they are “sticky” enough to grab onto molecules of interest. Northwestern’s Mirkin sees value in gold: specifically, nanoscale gold particles, to which he affixes multiple fragments of DNA that can latch onto DNA targets. Each gold particle becomes “like Velcro,” he says. In the next 18 months, Mirkin says, he and his colleagues will build a simple, doctor’s-office diagnostic device capable of instantly diagnosing diseases or predispositions to disease, depending on what DNA fragments are used on the device. “Chips will be built for panels of diseases,” says Mirkin, including sexually transmitted diseases, cystic fibrosis and genetic predispositions to colon cancer and blood hypercoagulation (blood that clots excessively).

Mirkin’s prototype chip, under development by Northbrook, IL-based Nanosphere, a company he cofounded, uses DNA deposited between electrodes on a microchip to recognize targets of interest. A sample is mixed with those “Velcro” gold particles and washed over the chip. If the sample contains the targeted DNA-say, genetic material from the syphilis bacterium-the DNA will bind to those sticky gold particles and then to the DNA fragments between the electrodes. The gold particles close the circuit and produce a detectable signal. The more electrode sensing elements per chip, the more diseases-or genetic predispositions-can be detected.

Mirkin’s group is adapting a process known as dip-pen nanolithography to gain the ability to literally “print” DNA molecules between electrodes just 200 nanometers apart. Mirkin hopes to pack hundreds, even thousands, of electrode sensing elements on one chip.

Printing Molecules In dip-pen nanolithography, molecules are printed directly on a chip surface.

Arrays of cantilevers (above) deposit millions, even billions, of different molecules on a surface; in cases where the printed molecules bind to specific genes or proteins, the chip can be used to diagnose diseases or discover drugs. Each cantilever, or “pen,” has a silicon tip (left) just a few atoms wide at its end. As the tip moves laterally, molecules attached to its sides are drawn down to the surface by a water meniscus that forms under the tip. The vertical motion of each cantilever is controlled thermally, allowing individual pens to start and stop printing.

Mirkin’s technology can find specifically targeted DNA in a sample. But if you could actually grab a single piece of DNA and directly “read” its genes, you could, in theory, identify any gene, or even complex gene patterns. Using tools adapted from semiconductor manufacture, physicist Harold Craighead of Cornell’s Center for Nanobiotechnology and his former postdoc Stephen Turner built a silicon chip containing tiny channels, each 50 nanometers in width and depth (see ” DNA Pipeline ,” below) . The channel is so small that a single strand of DNA can barely squeeze through-and that’s just the point. An electric field causes the normally coiled ball of DNA to bump into the channel, uncoil and thread its way down.

Once grabbed, the DNA needs to be “read”-to see, for example, if it contains a specific sequence. To make a sequence legible, researchers add fluorescent-labeled DNA probes to the sample beforehand; the probes bind to the target sequences. As each molecule of DNA wiggles its way down the channel, an optical detector identifies the fluorescent labels passing by. “We’re treating the DNA like it’s a recording medium,” says Turner, who is now president of Nanofluidics, a startup trying to commercialize the Cornell technology. “And just like a tape player, we’re playing the DNA.” While the Cornell researchers currently use an external optical microscope to read the “tape,” they hope to build an optical reader directly onto the chip using optical fibers. Turner expects to have a working device within the next few years.

Because the tools for making these tiny channels rely on the same standard equipment used to fabricate silicon chips for microelectronics, Turner envisions making nanofluidic chips with thousands and even millions of channels and optical fibers. With such devices, Turner says, doctors could one day take a drop of blood from a patient, drop it on the microchip and rapidly scan the DNA in the sample for genetic markers of disease. The device could also help doctors choose just the right drugs for the patient.

DNA Pipeline

To identify a particular sequence on a strand of DNA, researchers first mix the DNA with fluorescent probes that attach to that sequence. Then, on a microchip (above), an electrical field draws DNA through a channel 50 nanometers wide. An embedded optical reader detects any attached probes, identifying the sequence.

DNA Control

In the marriage of nanoelectronics and biology, the most extreme vision involves affixing electronic gadgets directly to molecules. To show how this might work-and why it might be useful-a team at MIT’s Media Lab, led by physicist Joseph Jacobson and biomedical engineer Shuguang Zhang, affixed gold particles, each only 1.4 nanometers in diameter, to a piece of DNA. Each gold particle served as a tiny antenna. The researchers then exposed the DNA to radio frequency magnetic fields, causing the particles to heat up, and the double-stranded DNA to break into two strands. When they removed the magnetic field, the strands came back together immediately. “Now we have a very powerful and useful tool that can control things at the molecular level,” says Zhang. “So far, there are no tools that can do this. To be able to control one individual molecule in a crowd of molecules is very valuable.”

That value, adds postdoc Kimberly Hamad-Schifferli, arises largely from the potential ability to turn genes on and off. To do that, the MIT researchers could attach fragments of DNA to gold particles. When added to a sample of DNA, the fragments would bind to complementary gene sequences, blocking the activity of those genes and effectively turning them off. Applying a magnetic field would then heat the gold particles, causing their attached fragments of DNA to detach, in effect turning the genes back on. Such a tool could give pharmaceutical researchers a way to simulate the effects of potential drugs, which also turn genes on and off. MIT recently licensed the technology to a biotech startup, Waltham, MA-based engeneOS.

Although remote control of DNA may sound more like a parlor trick than something your doctor might use, such experiments are demonstrating that nanoelectronics can interact with biology in powerful ways. Materials like nanowires and nanotubes, extensively researched by physicists and chemists in recent years, are now in the hands of biomedical engineers like MIT’s Zhang-with huge implications for everything from drug discovery to diagnosis of diseases like prostate cancer. While it’s difficult to predict winners among these many technologies, Berkeley’s Alivisatos, for one, says, “I think these things are all going to find competitive niches.”

Fast, cheap microelectronics revolutionized the world of computing and information technology. Whether nanoelectronics can revolutionize medicine remains uncertain. But the gap between electronics and biology is fast closing, and biomedical researchers and even physicians will soon have tools to probe life’s basic molecules in ways that seemed like fantasy just a few years ago.

Sensing Success
Some companies in nanobiotech

Doctors commonly diagnose infectious diseases by checking patients’ blood for evidence of proteins or genes unique to different bacteria and viruses. Soon, they may be able to look instead for pathogen-specific sugars, thanks to a glass chip developed by biologist Denong Wang at Columbia University’s Genome Center. The technology could ultimately be less cumbersome than DNA-based tests and more accurate than protein-based tests for certain pathogens, allowing physicians to quickly screen for thousands of different infectious diseases at once using a small sample of blood.

When a person is exposed to a bacterium or virus, his or her body produces antibodies that bind to specific sugar molecules on the pathogen’s surface. Wang dotted glass chips with some of those same sugars, from bacteria such as Pneumococcus or Haemophilus influenza. He then washed blood samples over the chips; if people had been exposed to Pneumococcus, for example, antibodies from their blood stuck to the corresponding sugars on the chips and were then detected through a microscope.

One of the biggest challenges in making sugar chips has been getting the sugars to stick to glass. Wang discovered a relatively simple method: he coated the chip surface with nitrocellulose, which holds the sugars in place. Each of Wang’s first chips contains just 48 different sugars, but, he says, “We could spot up to 20,000 different sugars on a single chip, which would allow us to target all the most common pathogens.” Wang’s goal, in fact, is to create a diagnostic tool that could detect pathogens such as HIV, anthrax and smallpox. He is in discussion with a number of drug companies about commercializing his technology.

Your dirt-biking expedition has ended painfully-a few ribs broken in a tumble on the trail-and the emergency-room doctor has sent you home with a bottle of codeine. It should be enough to tide you over until the bones heal, unless you’re one of the 20 million Americans who have a mutated form of an enzyme called cyp2d6, which normally converts codeine into the morphine that soothes pain. If you are, the enzyme won’t work, and the pills won’t even take the edge off. Worse yet, neither you nor your physician will know that until you take the drug.

Such is the reality of medicine today. Physicians can prescribe a drug based on a patient’s symptoms, but the hidden details of an individual’s genetic or molecular makeup can make him or her the wrong patient for that drug. Medications work differently in different people. What’s more, in the case of diseases like cancer or arthritis, a patient’s symptoms alone don’t always tell doctors exactly what’s wrong; subtle molecular differences can underlie seemingly similar illnesses. So choosing the treatment most likely to fix the problem is a hit-or-miss proposition. But that one-drug-fits-all reality is beginning to give way to a new era of “personalized medicine,” in which physicians can diagnose their patients with unprecedented accuracy and treat each of them with drugs tailored not only to the disease, but also to the patient’s genetic or metabolic profile.

“It’s going to totally transform medicine, there’s no question about it,” says Susan Lindquist, director of MIT’s Whitehead Institute for Biomedical Research. “And it’s going to be happening soon.” Mark Levin, CEO of Cambridge, MA-based Millennium Pharmaceuticals, offers one vision of what personalized medicine might mean for a patient: “When we walk into the doctor’s office 10 years from now, we’ll have our genome on a chip.” Using that chip, Levin says, a doctor will be able to determine what diseases a patient is predisposed to and what medicines will provide the most benefit with the fewest side effects. Even the way we think about disease will be different, says Jeffrey Augen, director of life sciences strategy at IBM, because doctors will make diagnoses based on genes and proteins rather than on symptoms or the subjective analysis of tissue samples under a microscope. “So instead of a person having chronic inflammation or cancer, he or she will have a cox-2 enzyme disorder or a specific set of genetic mutations,” Augen predicted at a recent conference in Boston.

The change is possible due in large part to emerging technologies that enable researchers to identify and analyze genes and proteins with phenomenal speed-thereby pinpointing the exact nature of different diseases and predicting individuals’ responses to drugs. Even using conventional DNA and protein analysis technologies, researchers have already taken some first steps toward personalized medicine. A woman with breast cancer, for example, can take a gene- or protein-based test that reveals whether her cancer will respond to certain drugs. But the key to gathering the massive amounts of genetic and molecular information that will expand personalized medicine’s reach-and make it a commonplace tool in the doctor’s office-is the thumbnail-sized biochip. These chips can analyze thousands of genes, proteins and other molecules at once from a single drop of blood.

One of the first triumphs for biochips in uncovering the molecular differences between diseases was a study led by biologists Patrick Brown at Stanford University and Louis Staudt at the National Cancer Institute in 2000. Using DNA microarrays-glass wafers spotted with thousands of DNA strands-the researchers examined patterns of gene activity underlying a type of cancer called non-Hodgkin’s lymphoma. After examining nearly 18,000 genes, they discovered that what was once thought to be one disease was in fact two distinct diseases. What’s more, the chemotherapy regimen normally prescribed for non-Hodgkin’s lymphoma patients was significantly less successful for patients with one of those two diseases-a clear indication that better knowledge of what’s going on at the genetic level could help doctors make better decisions about treatment.

DNA chips might soon begin to inform physicians’ decisions about how they prescribe some of the most commonly used pharmaceuticals. Santa Clara, CA-based Affymetrix and Basel, Switzerland-based Roche Diagnostics have teamed up to develop biochips that could help predict patients’ responses to such drugs as antidepressants and blood pressure regulators. The devices will be able to screen for several different mutations in the gene for the cyp2d6 enzyme-which helps metabolize a number of drugs in addition to codeine-and in another key enzyme gene. Roche aims to have the chips on the market by early 2003.

Even-more-sophisticated biochips might ultimately provide a quicker means of reading genetic fingerprints right in the doctor’s office. One drawback of existing DNA chips, for example, is that researchers first have to modify the sample of DNA in order for the chip to detect it. But physicist Scott Manalis and his group at MIT’s Media Laboratory are fabricating a silicon microchip that could potentially provide instant notification when it detects specific gene sequences in a sample of blood. In their device, micrometer-sized silicon cantilevers sense the molecular charges associated with biological molecules such as DNA and could produce a telltale electrical signal. “This opens up the possibility of making a simple biosensor for point-of-care diagnostics,” says Manalis.

Sometimes, however, DNA doesn’t tell the whole story. It’s often the proteins encoded by the DNA that actually determine whether a person is sick or well, and whether a drug is beneficial or toxic. Biologist David Sabatini at the Whitehead Institute found a way to look at the real-life activity of proteins by building arrays of living cells on glass chips. Sabatini recently cofounded the biotech firm Akceli in Cambridge, MA, to commercialize his technology, which he hopes to start selling to drug companies by mid-2003. Drug researchers could, for instance, equip each cell on the chip with a different variant of the body’s drug-metabolizing enzymes, and then expose the chip to a variety of drugs. By monitoring the cells’ responses, researchers could determine if a drug is toxic across the board, only to people with a particular enzyme variant, or not at all. “You can essentially create drug side-effect profiles,” says Sabatini. If a drug is toxic to some people but otherwise looks promising, a company may decide to pursue its development, targeting it to only those patients it benefits. Such a drug, developed specifically for people with not only a particular disease but a particular metabolic profile as well, would be the epitome of a personalized medication.

In the next decade, more and more such drugs, and the diagnostic tests necessary to choose among them, will begin to hit the market. So in the future, when you go to pop a pill you haven’t tried before, you won’t have to wonder if it’s really the right drug for you. You’ll know.

Alvaro Pascual-Leone holds a figure-eight-shaped paddle to his head and flips a switch. His left arm begins to twitch. He turns off the device-quelling its pulsing magnetic field, which was inducing an electrical current inside his brain-and his arm relaxes. But Pascual-Leone, a neuroscientist at Boston’s Beth Israel Deaconess Medical Center, is interested in more than muscle twitches; he believes that magnetic stimulation provides the last, best hope for treating patients with severe depression. This fall, researchers will begin large-scale human trials of the technology to see if he is right.

Some 20 million Americans suffer from severe depression, and many of them don’t respond to conventional drugs. Based on several small studies, including Pascual-Leone’s, regulators in Israel, Europe and Canada gave magnetic stimulation the green light last year as an alternative depression treatment. With a version of the technology licensed from Emory University, Atlanta-based Neuronetics plans to begin human trials, involving 240 patients at eight sites around the country, in September. If all goes well, Neuronetics’ CEO Stan Miller hopes to see U.S. psychiatrists offering magnetic treatments for severe depression by 2004.

Mark George, a neuroscientist at the Medical University of South Carolina participating in the trials, sees magnetic stimulation as an alternative to shock therapy. “While shock therapy is very effective, it has lots of side effects such as memory loss, and you have to give the patient a seizure,” says George. Magnetic stimulation has none of those side effects. The technique excites nerve cells in a small area of the brain that is underactive in depressed patients.

Still, some believe that the move to the clinic may be premature. “I think the therapy has gotten way ahead of the science,” says National Institutes of Health neuroscientist Eric Wasserman. But the reality, say other researchers, is that many depressed patients are desperate for alternative treatments and would gladly try magnetic stimulation despite uncertainties.

While microchip makers continue to wring more and more from silicon, the most dramatic improvements in the electronics industry could come from an entirely different material: plastic. Labs around the world are working on integrated circuits, displays for handheld devices and even solar cells that rely on electrically conducting polymers-not silicon-for cheap and flexible electronic components. Now two of the world’s leading chip makers are racing to develop new stock for this plastic microelectronic arsenal: polymer memory.

Advanced Micro Devices of Sunnyvale, CA, is working with Coatue, a startup in Woburn, MA, to develop chips that store data in polymers rather than silicon. The technology, according to Coatue CEO Andrew Perlman, could lead to a cheaper and denser alternative to flash memory chips-the type of memory used in digital cameras and MP3 players.Meanwhile, Intel is collaborating with Thin Film Technologies in Linkping, Sweden, on a similar highcapacity polymer memory.

Polymer microelectronics are potentially far less expensive to make than silicon devices. Instead of multibillion-dollar fabrication equipment that etches circuitry onto a silicon wafer, manufacturers could eventually use ink-jet printers to spray liquid-polymer circuits onto a surface. Polymer memory comes with an added bonus: unlike the memory in your PC, it retains information even after the power is shut off. Such nonvolatile memory offers potential advantages- not the least of which is the prospect of never having to wait around for a PC to boot up-and a number of researchers are working on various approaches (see “Magnetic Random-Access Memory,” TR July/August 2002). But polymer memory could potentially store far more data than other nonvolatile alternatives.

Polymer memory stores information in an entirely different manner than silicon devices. Rather than encoding zeroes and ones as the amount of charge stored in a cell, Coatue’s chips store data based on the polymer’s electrical resistance. Using technology licensed from the University of California, Los Angeles, and the Russian Academy of Sciences in Novosibirsk, Coatue fabricates each memory cell as a polymer sandwiched between two electrodes. Application of an electric field to a cell lowers the polymer’s resistance, thus increasing its ability to conduct current; the polymer maintains its state until a field of opposite polarity is applied to raise its resistance back to its original level. The different conductivity states represent bits of information.

Coatue’s polymer memory cells are about one-quarter the size of conventional silicon cells. And unlike silicon devices, the polymer cells can be stacked to produce a three-dimensional structure. That architecture could translate into memory chips with several times the storage capacity of flash memory. By 2004, Coatue hopes to have memory chips on the market that can store 32 gigabits, outperforming flash memory, which should hold about two gigabits by then.

But turning polymer memory into a commercial product won’t be easy. Memory technologies compete not only on storage capacity but on speed, energy consumption and reliability. “The difficulty is in meeting all the requirements of current silicon memory chips,” says Thomas Theis, director of physical sciences at IBM’s Watson Research Center in Yorktown Heights, NY. Until new memory materials are able to compete with the high performance of silicon, Theis notes, they are likely to be limited to niche applications.

One likely use is in disposable electronics, where cost, rather than performance, is the deciding factor. Researchers at Lucent Technologies’ Bell Laboratories are working on polymer memory devices for use in identification tags. The polymer memory made at Bell Labs is still relatively slow by silicon standards, and anticipated capacity is only on the order of a kilobit. But, says Bell Labs chemist Howard Katz, the flexible and low-cost polymer memory devices could be “very attractive” for, say, identification tags meant to be thrown away after a few uses.

As polymer memory technology advances, it could pave the way to computers made entirely of plastic electronic components, from the display to the logic chip. That may be decades off, but as researchers push the bounds of polymers, the vision seems less far-fetched. And in the short term, Coatue says its polymer memory could be integrated into the existing silicon infrastructure. “The revolution has already begun,” says MIT chemist Tim Swager, a scientific advisor to Coatue.

The main skill in diagnosing neurological disorders such as multiple sclerosis and Alzheimer’s disease: educated guesswork. Indeed, today’s doctors rely primarily on interviews, physical examination and laboratory tests to detect these complex neurological diseases; the problem is that symptoms can vary dramatically from one patient to the next, making diagnosis tricky and subjective. But by combining new databases with improved medical-imaging techniques able to resolve telltale anatomical features a millimeter in size or less, researchers are starting to make the invisible visible, potentially enabling them to offer patients earlier and more accurate diagnoses.

At the State University of New York at Buffalo, for example, researchers have developed software that renders three-dimensional pictures of the brain from magnetic-resonance imaging data, allowing them to digitally parcel off areas of the brain and precisely calculate their size and volume. Rohit Bakshi, director of the Buffalo Neuroimaging Analysis Center, has used the technology to show that the caudate nucleus-a part of the brain’s gray matter involved in motor control and thinking-is significantly smaller in multiple-sclerosis patients than in healthy patients (see image). Through such software tools, Bakshi hopes to standardize the way neurologists analyze MRIs. “Today, two clinicians can look at the same MRI and see it differently,” says Bakshi. “We’re working on making MRI a quantitative and standardized test, like a blood test, where you get a specific, reliable value back, and you can accurately compare the results to normal people.”

Besides aiding diagnosis, the new techniques could help track the course of a disease-and the benefits of treatments. Bruce Rosen, director of the Martinos Center for Biomedical Imaging at the Massachusetts General Hospital in Boston, and his colleagues are already using MRI machines to measure the thickness of brain structures only one-tenth to two-tenths of a millimeter in size. “That means we could see changes in the brain in response to a drug that occur in three to six months instead of assessing memory improvements, which tend to evolve over 12 to 18 months,” says Rosen.

To help standardize the diagnosis process, Bakshi’s center is developing a large database of brain scans taken from multiple sites across the state of New York. With more than a thousand images already in stock, he and his colleagues are building software that correlates scans of multiple-sclerosis patients with data about the courses the disease takes with them, to identify variations and predict whether patients will recover or develop chronic illness. A consortium of U.S. universities, which includes Rosen’s imaging center, started work this year on a similar database network containing brain scans of Alzheimer’s patients from across the country. The $20 million project will connect databases at hospitals and universities in California, North Carolina and Massachusetts. “Ultimately, our hope is that when somebody comes in and we take a brain scan, we can make a diagnosis, stratify their disease and determine what treatments would be most effective,” says Rosen.

“Databases like these are definitely the future,” says Robert Knowlton, a neurologist at the University of Alabama at Birmingham Hospital. Better classifications of brain diseases, he says, will ultimately lead to better treatments.

The new NanoMechanical Technology Laboratory brings a burgeoning field closer to the real world.