| The Fashion Issue | winter 2007 |
Boutique Medicine Stroke patients may soon be wearing Italian-designed gloves as part of their With shrinking circuitry, burgeoning wireless technology, and more powerful and specific algorithms, wearable systems that monitor movement, mood, and muscle activity are poised to go from the laboratory to the clinic and the home. Harvard researchers are now developing such devices to facilitate the monitoring and treatment of patients with a range of conditions, including depression, stroke, and Parkinson’s disease. “Most doctor–patient interactions take place on an outpatient basis, allowing the physician only a sliver of time with the patient,” says Carl Marci ’97, director of social neuroscience at Massachusetts General Hospital. “Part of what’s driving the wave of wearable technology is our desire to monitor patients continuously. When we understand more fully each patient’s condition on an hour-to-hour basis, we can create better models of care.” Wired for Sound Efforts to monitor a patient’s condition have been around for a long time. It was more than a century ago, for example, that Willem Einthoven invented the mechanism of the electrocardiogram, or ECG. Recent advances have shrunk the size of such devices while computing power has expanded their potential applications. The result has been instruments such as the ring sensor, which encircles a patient’s finger to monitor heart rate, the variability of that rate, blood pressure, and oxygen saturation. A similar wireless system—dubbed CodeBlue—is currently being developed for use at Beth Israel Deaconess Medical Center by Matt Welsh, an assistant professor of computer science in the School of Engineering and Applied Sciences at Harvard University. Welsh is designing an integrated wireless technology system that will continuously—and unobtrusively—monitor patients waiting in the medical center’s emergency department. This iPod-sized device can be worn around the neck on a lanyard. Wired to a pulse oximeter and two ECG leads, the device also features an LCD screen, a button for alerting medical staff in case of emergency, and a pager that allows staff to contact the patient. Welsh is also applying wireless technology to wearable devices for stroke patients and for those with Parkinson’s disease. The devices are about the size of the key fobs used as remote controls for car-door locks. But Welsh expects future models to be small enough to be embedded in articles of clothing or fastened to straps. Each sensor consists of a microcontroller, a low-power radio to transmit data, and an accelerometer and gyroscope to capture a patient’s limb movements. Intel Corporation’s Digital Health Group in Cambridge, Massachusetts, developed the platform. Welsh’s group wrote the application software running on the sensor, and a research group led by Paolo Bonato, director of the Motion Analysis Laboratory in the Department of Physical Medicine and Rehabilitation at Spaulding Rehabilitation Hospital, has been developing algorithms that will capture and analyze the data. Bonato’s group also tests the system and designs different types of wearable devices to suit a range of applications. Rocket science “In designing wearable technologies, we always start with the specific challenges of the clinical problem,” says Bonato. In one study, Bonato’s team asked a dozen people with Parkinson’s disease to perform various tasks while wearing sensors attached to their bodies. Patterns of movement were recorded using the same technology installed on missiles to track their trajectories. The investigators then analyzed the data for movement patterns associated with various severities of symptoms. Each analysis produced a patient-specific score on the Unified Parkinson’s Disease Rating Scale. Bonato found that the approach could predict each patient’s level of dyskinesia—an impairment of the ability to control movements—and bradykinesia—a slowing of body movement. He considers the study one step toward improving the wireless system’s functions while also refining the system’s design to allow patients greater freedom of motion. “Managing the symptoms of Parkinson’s patients is challenging,” says Bonato, “because they constantly evolve. If you adjust a patient’s medication, you want to know whether that change is working. It’s difficult, though, for patients to report changes in their motor skills objectively. Wearable technology can provide objective measures on a continuous basis. The patient could strap elastic bands with wireless components around the limb that would be most affected by the medication change. The sensors could capture information for a week or so, after which the physician could have the data analyzed. You’d then have a detailed picture of whether the medication adjustment was effective.” Researchers must worry about more than fine-tuning the system and its analytic capabilities; its expense must be tweaked as well. Each prototype now costs approximately $350, but scaling up production would drop the price per unit considerably. “If I were to build an iPod on my own, it would cost me thousands of dollars,” Welsh says, highlighting a technology for which the wedding of miniaturization and mobility produced new applications. “But if we were making 10,000 wearable sensors, the cost would fall to about $100 each. Some believe it could ultimately drop to $20.” REALITY CHECK For patients recovering from stroke, Bonato studies wearable sensors that could be used to tailor rehabilitation exercises, test the efficacy of different forms of rehabilitation, and determine whether improvements measured in a hospital setting carry over to people’s daily lives. For this work, Bonato partners with Joel Stein, chief medical officer at Spaulding Rehabilitation Hospital. “In the lab, you may be able to demonstrate that patients can move their stroke-weakened arms better after rehabilitation,” says Stein. “But when they return home, they may avoid using their weakened arms. Our goal is to extract real-world knowledge of how people are using their muscles.” In the ongoing pilot study, the researchers aim to create better sensor arrays and to validate the algorithms that make sense of the data collected. The study participants have sensors with accelerometers attached to their stroke-affected arms. Measurements are taken during a series of tasks, such as drinking from a can and pushing and pulling a weight across a table. Stein says that more sophisticated methods of measuring functionality are needed to judge the effectiveness of newer therapies being applied to stroke rehabilitation, such as robotics, growth factors, stem cells, and brain stimulation. Other researchers seek to augment patient health through wardrobe accessories, in one case, a “smart” shoe that is both fashionable and musical. Designed to help the elderly, stroke patients, and people with Parkinson’s disease retune their gait, the shoe has been tested at Massachusetts General Hospital in research involving a small group of Parkinson’s disease patients. The footwear combines thin, pressure-sensor insoles with accelerometers and gyroscopes that are attached to the heels of the patients’ own shoes, says Donna Moxley Scarborough, a physical therapist at the hospital and a member of the research team. The sensors work together to measure gait and transmit data wirelessly to a computer, where it is analyzed and fed into a synthesizer. Music is then matched to gait and paced at a tempo researchers hope will, through biofeedback, help the patient become more sure-footed. Clinically depressed patients also may benefit from wearable monitoring. One study Marci led at Massachusetts General Hospital outfitted such patients with an arsenal of sensors. These included an accelerometer; leads to measure a patient’s heart rate, skin conductance, body motion, and location; and a microphone to capture speech patterns—all built to transmit data through Bluetooth technology. “What really surprised us about some of these features was how well they enhanced our ability to predict where a patient would place on the Hamilton Depression Rating Scale and even that patient’s overall ability to function,” says Marci. He envisions using sensor-based monitoring for decisions on how to tailor drug dosage and to adjust treatment as a patient’s condition changes, much as blood pressure checks now allow physicians to better calibrate treatments for hypertensive patients. fashion therapy More recently, Bonato’s research has begun to blur the line between medical device and clothing. “If the physician needed to monitor the movement of multiple body regions, the patient might have to strap on numerous sensors every day for months,” Bonato explains. “After a few weeks, that patient would tire of the routine, and compliance would disappear.” So Bonato’s team is working with materials that have conductive elastomers printed onto the fabric—to eliminate the need to attach wires under a person’s clothing—as well as printed on gloves to monitor hand movements. Data-gathering shirts could track a patient’s upper body movement, while sensor-enhanced gloves worn during stroke rehabilitation could both measure hand motion and act as an interface between a patient and a real-time video system that displays the patient’s movements on screen. Some high-end sports gear manufacturers, such as those serving the training needs of competitive athletes, have already created commercial products in the form of machine-washable shirts that measure heart rate, respiration, and posture. Bonato also hopes to weave wireless technology into textiles. He is working with researchers in Italy—at the University of Pisa’s bioengineering, biorobotics, and artificial intelligence laboratory and at a company, Smartex—to develop prototypes of such “smart” garments. Bonato speculates that using such material to make everyday clothing would not only enhance data-gathering possibilities but also increase patient compliance. “The only drawback might be that a shirt would not be in a style the patient would want,” says Bonato. “Smartex is looking into different designs to allow patients a range of style options.” The gloves, he adds, offer several advantages over current monitoring systems. At $10 apiece, they are significantly cheaper than equivalent virtual-reality systems, which cost upwards of $30,000. “You don’t need an expensive platform,” Bonato says. “Using spandex, you can produce gloves that fit perfectly, are machine washable, and are so inexpensive that each patient can have a pair throughout rehab.” As for fashion, the gloves can be ordered in an array of colors. And the glove designer’s other work? Designing for the fashion house Dolce & Gabbana. Ken Wilan is a freelance writer based in Westborough, Massachusetts. Photo: ©2007 Jupiterimages Corporation |
rehabilitation. And people with Parkinson’s disease may be selecting an undershirt that confers benefits well beyond those offered by Fruit of the Loom. The next wave of wearable monitoring systems is proving smarter than ever, as form and function merge to produce designs for better living. 
|