Discuss This Article

On the Front Lines of Telemedicine

By Michael W. Dorsey

In a bomb-scarred building, on a dusty, rubble-strewn street in a foreign city, a U.S. soldier is in mortal trouble. Nearby, an Army medic huddles in a doorway, a tiny computer screen set in front of his eye like a jeweler's loupe. The numbers on the screen tell the soldier's story: blood pressure plummeting, pulse slowing, blood oxygen dropping.

Tiny wireless sensors attached to the soldier's body monitor his failing vital signs and radio them to a pager-sized transmitter strapped to his belt. The transmitter encrypts the information and broadcasts it--along with the soldier's exact location in three dimensions--to the medic, who is soon at his side.

Unzipping a pouch in his jacket, the medic pulls out an ultrasound transducer the size of a computer mouse and switches on his small, wearable computer. "Scan," he calls into a helmet-mounted microphone. As he probes the soldier's abdomen, an image flashes on his eyepiece revealing internal injuries from an AK-47 round.

The medic radios for help and broadcasts the soldier's ultrasound images to the field hospital. All the while, the sensors keep hospital personnel posted, moment by precious moment, on the state of the soldier's health. When the medevac chopper touches down, surgeons are standing by, armed with the information they need to immediately work to save the young man's life.

Building on a Solid Foundation

This is the vision of the future of battlefield medicine that WPI is helping to create in its Center for Untethered Healthcare. It builds upon more than a decade of work on wireless networking, noninvasive medical sensors and ultrasound imaging. Providing critical medical data to medical personnel where and when they need it will increase the odds of survival for the wounded or injured.

Cadet Erica Schmidt, a senior at Assumption College in Worcester, is training to be a medic with the Bay State Battalion of the Reserve Officers' Training Corps (ROTC), headquartered at WPI. She represents the next generation of military medics who will be able to monitor an entire cadre of troops from afar using wearable sensors, data transmission networks, and portable ultrasound technologies being developed at WPI in conjunction with the U.S. Army. If history is a guide, these advances in military medicine will find their way into the civilian sector, improving healthcare for all of us.

Congress appropriated an initial award of more than $800,000, through the U.S. Army Medical Research and Materiel Command (USAMRMC) at Fort Detrick, Md., for creating technology to monitor the health of soldiers in the field in real time. An additional $1 million was appropriated for the project in FY03.

The center is one of four research entities that make up the university's new Bioengineering Institute (BEI). Headquartered at Gateway Park, a 10-acre industrial area a few blocks from campus (and thousands of miles from any hot spot) being redeveloped by a for-profit partnership of WPI and Worcester Business Development Corporation, BEI fosters research in untethered healthcare, bioprocess and tissue engineering, molecular engineering, and comparative neuroimaging. It's designed as an incubator for startup companies and will also license technology to established biomedical and pharmaceutical firms. Current corporate partners include Abbott Laboratories, maker of healthcare products from antibiotics to nutritional drinks, and Nypro, a leading injection-molding firm in Clinton, Mass., specializing in bioengineered products.

WPI's foray into untethered medicine brings together three mature lines of research in two academic departments: noninvasive physiological sensors, a longtime focus of research for Yitzhak Mendelson, professor of biomedical engineering; wireless communications and geolocation, the specialty of William Michalson, professor of electrical and computer engineering and director of the Center for Untethered Healthcare; and advanced techniques for medical ultrasound, the work to which Peder Pedersen, professor of electrical and computer engineering, has devoted the past 15 years.

Separately, Mendelson, Michalson and Pedersen must overcome a host of technical obstacles to complete the portion of the system they are developing for the Army. But each will also face a number of common challenges. Many of these are directly related to the fact that their technology must be carried into the field and used by soldiers and medics under hostile conditions.

Advanced Physiological Sensors
Small, intelligent wireless sensors will monitor vital signs of soldiers in real time, alerting medics and field commanders when problems arise. Yitzhak Mendelson is developing sensors that will measure pulse rate, skin temperature and blood oxygenation. To extend battery life, the sensors will use low-power LEDs surrounded by a ring of highly sensitive light detectors.

For example, electronic gear designed for field use must be rugged and reliable. Soldiers already carry up to 90 pounds of equipment and supplies, so it needs to be lightweight. Since it will be employed amid the chaos of war, it must be easy to use. One of the most important and vexing challenges the researchers will face is minimizing the power requirements of their systems.

"Soldiers are already equipped with all sorts of devices, including radios, that require batteries," Michalson says. "Batteries are now a soldier's lifeline to the outside world. Between 30 and 50 percent of the weight a soldier carries consists of batteries. In fact, I've heard from soldiers who say they'll get rid of clothing and food so they can carry more batteries."

To minimize power use, Michalson, Mendelson and Pedersen will likely give their devices standby and sleep modes that reduce power needs to the bare minimum. Mendelson says he hopes to create intelligent sensors that remain silent unless a medic calls for a reading or until an anomalous reading is detected.

Keeping Tabs on Vital Signs

In his laboratory, Yitzhak Mendelson peers through a huge magnifier as he assembles the tiny components of sensor prototypes with the concentration of a watchmaker. With eight patents to his credit, Mendelson is a keen innovator. He is developing the wireless sensor to monitor pulse rate, skin temperature and arterial oxygen saturation--a measure of how fully charged red blood cells are with oxygen. Blood loss or injury to the lungs would cause saturation to dip.

To understand the state of a patient's health, a doctor gathers basic data such as heart and breathing rates, temperature, and blood pressure. Through its Warfighter Physiological Status Monitoring program, the U.S. Army hopes medics and field commanders can keep tabs on the vital signs of every soldier by way of wireless sensors attached to a soldier's body or built into his uniform.

Wearable Ultrasound Scanners
Starting with off-the-shelf technology, Peder Pedersen will develop an ultrasound unit built around a wearable PC. Medics will operate the unit with voice commands, to keep their hands free, and view images in a flip-down eyepiece. Pedersen will also tackle the daunting challenge of developing techniques to process images of injuires and wounds to make them easier to interpret.

Mendelson has been working for more than a decade to advance the technology for measuring oxygen saturation with a technique known as pulse oximetry. Pulse oximeters shine light of two specific frequencies through the fingertip or earlobe and then measure the intensity of the light transmitted to a photo-detector. The technique is based on the knowledge that well-oxygenated blood is bright red, while oxygen-poor blood is a darker, bluish red.

One of his innovations was to place the oximeter's light-emitting diodes and photodetector side by side. Since such a sensor measures reflected light, rather than transmitted light, it can be placed almost anywhere on the body (readings from peripheral areas like the fingertips and ears can be unreliable in cold weather or when the body has lost a lot of blood). Using this technique, Mendelson is developing a sensor that can be applied to a fetus to monitor oxygen saturation in real time during labor and delivery.

Among the challenges Mendelson will face are making the sensors as small and light as possible, and building in circuitry for power management and advanced signal processing. He must also devise a way to keep the devices in contact with the soldier's skin, no matter how sweaty or grimy.

"Surprisingly, this will be one of the more difficult challenges," he notes. "It will take some research to determine how best to keep the sensors in place where they can do their job and still make them relatively unobtrusive to soldiers." Various types of tape and adhesive, sensors built into clothing or the headband of a helmet, and sensors that double as rings will be among the options studied.

Power use will be a critical issue, as well. The most power-hungry components of the sensor will be the light-emitting diodes. The bright red and infrared light they produce also concerns the Army, since it could give away a soldier's position at night. Mendelson will likely address both issues by using low-power diodes that emit little light. He'll surround them with a ring of detectors capable of capturing the small amount of reflected light.

Giving Medics Inside Knowledge

Two soldiers lie wounded, but only one can be evacuated right away. How can a medic know which one has massive internal bleeding and which one took a bullet that miraculously left his vital organs untouched? "Ultrasound imaging technology can provide this kind of information, which cannot be obtained in any other way," notes Peder Pedersen. "It's not feasible to take X-rays or MRI scans in the field, so the best choice is ultrasound."

To develop an ultrasound unit for the Army, Pedersen will begin with existing hardware and software, including a wearable personal computer and a Terason 2000 portable ultrasound scanner from Teratech Corp. The Terason is the only portable ultrasound unit currently on the market that runs on a regular PC, which will enable Pedersen to add his own enhancements.

Those add-ons will include power management, image enhancement and voice recognition software. Medics need to have their hands free (one to hold the transducer and one to support the patient), and bringing a computer monitor into the field is impractical. Plans call for operating the scanner with voice commands, rather than a keyboard or a mouse, and viewing images on a flip-down eyepiece. Power management software will extend battery life while assuring that the scanner and PC are available at a moment's notice.

Wireless Networking and Location
William Michalson is developing the wireless protocols and signals that will permit sensor data and ultrasound images to be transmitted reliably in the unforgiving environment of the urban battleground. The signals must be encrypted, be difficult to jam, and support hundreds of users in a relatively small area. Michalson will also build in technology that will transmit a soldier's exact location.

Pedersen's tasks include adapting existing voice recognition software by developing a small vocabulary of simple, distinctive and easy-to-remember commands. He'll incorporate signal processing algorithms that will enable the software to filter out background noise--whether thumping helicopter blades or gunfire. Developing this hardware and software will take time, but Pedersen says his greatest challenge will be finding ways to display images of injuries so that medics, who are not likely to have had extensive training in ultrasound, can readily decipher them.

"We want to help the medic make the right decisions for critical injuries," Pedersen says. "We're not talking about using the system to see subtle things. We're looking to be able to determine the extent of bleeding and internal injuries. How do we present those images so they are easy to interpret?"

Pedersen and his students will simulate common internal injuries using tissue phantoms--materials that look to an ultrasound scanner like organs, blood vessels and other tissues. Next they'll develop intelligent image processing techniques that can recognize and enhance these specific images.

In this part of the project, Pedersen will draw on his extensive work with modeling on a computer how the signals that produce two-dimensional ultrasound images are generated from reflections off three-dimensional tissue and organ structures.

"From fairly simple experimentation to the complex structure of the human body is a big jump," he notes. "There is a lot of research between where we are and where we want to be, making it difficult to say just how successful we will be."

Making the Right Connections

A soldier crouches in an alleyway, waiting for the enemy who could be anywhere: around the block, on the tenth floor of a nearby building, in the next alley.

"Today's battlefield is more likely to be an urban environment--fighting building to building, floor to floor," says Bill Michalson, who is working on a wireless link to transmit images and data back to medics. "It's a horrible situation for wireless."

The third part of WPI's contribution to future battlefield medical systems is this wireless link. While it may seem like the most straightforward part of the project, it is one of the most complex, according to Michalson.

Wireless protocols employed in today's cell phones and wireless networks are inadequate for use in combat, Michalson says. "They're not secure, they're too easy to detect, and they can be easily jammed. They're not designed to work in the highly complex environment of the modern battlefield."

Michalson is approaching this challenge by studying existing wireless protocols under realistic conditions to better understand their strengths and weaknesses, and by focusing on the design of signals, or waveforms, that exhibit specific characteristics (difficult to detect, high bandwidth, etc.) with the hope of finding one that meets the Army's daunting requirements.

"It has to support as many users as possible in a confined space, have properties that make it stealthy and hard to jam, be able to support transmissions at the kind of bandwidth we need, and be effective in the indoor environment. It's a massive challenge, and some of the characteristics are mutually exclusive."

The wireless systems Michalson will develop must not only send and receive communications, but transmit the exact location of each soldier. This portion of the project will draw on Michalson's extensive work on using the Global Positioning

System in various transportation applications, some of which was sponsored by the Federal Aviation Administration and the U.S. Forestry Service. In particular, it will continue a line of work that started with an undergraduate project in 1996.

ECE majors Michael Roberts, William Cidela and Chris Mangiarelli developed technology to alert engineers on Providence & Worcester Railroad locomotives when they were approaching a switch set in the wrong position. As one of the advisors for that project, Michalson became intrigued with the challenge of locating an object, such as a locomotive, as it moved through confined spaces like tunnels and rock cuts.

This interest led to further discussions with representatives of the railroad and mining industries. After the 1999 Worcester Cold Storage warehouse fire, remembered for desperate efforts to locate six firefighters who ultimately died inside the blazing building, Michalson was part of a WPI team that proposed a system for monitoring and locating emergency workers in buildings. (The project recently received a $1 million appropriation from Congress; see page 6).

"Whether we're designing systems for firefighters and rescue workers, military personnel, or hospitals--they'll all use the same fundamental signal design," says Michalson. "The system requirements of the military and civilian sectors may be quite different, but the commonality is the signal design. So, the benefits in one area will be helpful to the other areas."

From the Front Line to the Home Front

Translating the benefits of this medical technology for the military into the civilian heathcare realm is one of the most important missions of BEI, according to director,Tim Gerrity.

"The biggest burden on our healthcare system is the chronically ill," Gerrity notes. "Asthma affects about 10 percent of all Americans; diabetes and arthritis each affect about six percent. So you have a large portion of the population suffering from a chronic illness that requires some level of management on a day-to-day basis. At the same time, you have healthcare costs that are rapidly escalating--to the point where some major companies are finding it necessary to reduce or eliminate health benefits for employees."

Smart physiological sensors, in concert with a wireless network, could permit elderly or chronically ill patients to be monitored in their homes, eliminating the need for frequent trips to the doctor and making physicians more productive. More important, such sensor systems can monitor trends and detect anomalies, so that required treatment, including emergency treatment, can be anticipated and delivered just in time.

"By reducing the need for office visits, and all the time and effort it takes to schedule them," Gerrity says, "and by giving patients real-time information they can use to better manage their own health, untethered healthcare has the potential to significantly reduce the cost of delivering healthcare while also increasing the quality of care people receive."

It will be three to five years before any civilian applica-tions begin to hit the market. But when the time comes, Gerrity says BEI will do everything it can to make it a successful transition. The institute's mission, he says, is to take technology developed by its faculty and students and commercialize it through corporate partnerships and by nurturing the startup companies.

"You wouldn't believe the amount of technology that has been developed in this field that ends up gathering dust on a shelf somewhere," he says. "I can assure you that isn't going to happen with the work of this center."

Maintained by: webmaster@wpi.edu
Last modified: Sep 02, 2004, 12:01 EDT
[WPI] [Alumni] [Home]