An arms race
Building better prosthetic limbs
As you hold this magazine, you’re probably not thinking about the thousands of electrical signals traveling between your fingertips and your brain. These signals control your hands’ grip on the magazine and produce motions to adjust it to the right angle or to turn the page (though hopefully you’re not ready to do that just yet). They also provide an equally rich amount of data back to your brain that constantly refines your movement and allows you to feel the weight and temperature of the paper. These very same electrical signals may someday help arm amputees control their prosthetic hands and arms just as you control yours — without even thinking about it.
Researchers at McCormick, in collaboration with members of Northwestern’s Feinberg School of Medicine and the Rehabilitation Institute of Chicago (RIC), are exploring ways to build and control better prostheses for arm amputees. Some truly revolutionary breakthroughs here have given hope to people who have lost limbs and are playing a major role in the rapid advancement of prosthetic technology.
The need for improved prostheses is driving a $48.5 million national push from the Defense Advanced Research Projects Agency (DARPA). Todd Kuiken (PhD ’89, Feinberg ’90), associate professor of biomedical engineering and physical medicine and rehabilitation and director of the Neural Engineering Center for Artificial Limbs at RIC; Richard Weir (MS ’89, PhD ’95), clinical professor of biomedical engineering and research associate professor of physical medicine and rehabilitation; Michael Peshkin, professor of mechanical engineering; and Ed Colgate, Pentair–D. Eugene and Bonnie L. Nugent Teaching Professor and professor of mechanical engineering, are pursuing collaborative research that aims to speed the development of a lifelike prosthetic arm and hand.
Engineering from both sides
It’s one of the first rules of engineering design: design with the user in mind. But what if you can also modify the user to better interact with the design? That’s the key concept in Todd Kuiken’s work in nerve transfer, which began almost 20 years ago when he was a PhD student at McCormick.
Kuiken takes the nerves that would have gone to a missing arm and transfers them into the pectoral muscle, which no longer has a purpose once the arm is gone. The nerves then grow into the new muscle, telling it to contract and relax based on the signals that would have controlled the missing arm. The signals from nerves that have not been transferred are too small to measure reliably for long periods of time. By rerouting the nerve endings into the pectoral muscle, the signals become stronger, and Kuiken and his team are able to use sensors to detect them. Using these signals, the patient is able to intuitively control a motorized prosthetic arm. A patient’s thought to “close my hand” becomes a command to close the prosthetic hand.
“This process has a number of advantages,” explains Kuiken, “the biggest being that you use muscle as a biological amplifier as opposed to relying on hardware. It never breaks down and has an infinite energy supply — as long as you eat your Wheaties.”
Kuiken’s first challenge was to decode the meaning of the electrical signals, differentiating the signals that cause the hand to open from those that cause it to close. One of his first collaborators at McCormick was Allen Taflove, professor of electrical engineering and computer science. “I went to Allen and told him I was interested in using finite- element modeling and wanted his help,” Kuiken says. “He was wonderful in helping me to get started, figure out the modeling, and get our first grant. He was so generous with his time, support, and enthusiasm during a very critical time.”
Kuiken spent several years working through decoding problems and doing the bench work and animal studies necessary to prove the feasibility of his concept. Ready to begin clinical trials, Kuiken identified his first patient: Jesse Sullivan, an electrical line worker who burned his arms so severely that both were amputated at the shoulder. Following his successful nerve transfer in 2003 and the fitting and implementation of new prosthetic arms, Sullivan has been dubbed the world’s first “bionic man” and serves as a living example of the promise of this technology.
After transferring four of Sullivan’s nerves into his pectoral muscle, Kuiken had modest goals: to allow Sullivan to open and close his arm and bend and straighten his elbow in a natural way. Using sensors that picked up the electrical signals rerouted to Sullivan’s chest to operate the three-motor prosthetic arm, Sullivan was able
to control his arm so well that Kuiken set his sights even higher.
Working with Richard Weir — whom Kuiken met when they were both PhD students at McCormick — and other collaborators from around the world, Kuiken developed a six-motor arm that provided six degrees of freedom. This was a vast improvement over the three-motor arm but still fell far short of the 22 degrees of freedom in a human arm. Cobbling together an elbow from Boston, a shoulder from Scotland, a hand from China, a rotator from Vienna, and humeral rotator from Weir’s lab, Kuiken and Weir created a new arm with twice the functionality of Kuiken’s original model.
“In the first two weeks, Jesse did remarkably,” Kuiken says. “He’s an absolutely wonderful guy to work with. Those results got us going and got us excited.”
While nurses prepped Jesse Sullivan’s chest with rubbing alcohol during one of his many visits to RIC, a remarkable thing happened: Sullivan felt the cooling effect of the alcohol as though it were on his hand. Searching for an explanation for this phenomenon, Kuiken discovered that the nerves transferred into Sullivan’s chest actually grew into the skin on his chest, a process known as targeted sensory reinnervation. Both the outgoing and incoming nerve signals for the arms had regrown into the pectoral muscle and skin. With this unexpected finding, Kuiken saw even greater potential. “This gives you a portal to let the person feel what they touch as if it were in their missing hand,” he says.
While Kuiken was excited about this advance, he knew that he didn’t have the expertise to put it to work. Richard Weir and Jon Sensinger, a PhD student in biomedical engineering, worked to develop a proof of concept for a device that could communicate the sense of touch to a patient’s chest. After seeing that it could work, the team connected with Michael Peshkin and Ed Colgate in the mechanical engineering department at McCormick, who have based a significant part of their research on the study of haptics, or the sense of touch, mostly in relation to robotics.
Colgate and Peshkin are now developing tactors — microrobots that can convey haptic sensations to a patient’s chest — for use in conjunction with prosthetic arms. Using inputs from the prosthetic arm, these tactors recreate the same sensation on a scale appropriate to the area of skin on the chest where the nerves have reattached.
While the research is still in its early phases, the results have been remarkable. The device can apply force in several directions and even heat up and cool down based on temperature sensors. In testing, Sullivan and Claudia Mitchell, a single-arm amputee who became the first woman to be outfitted with the bionic arm, have been able to differentiate between satin ribbon and sandpaper and feel temperature changes. This advance has both practical and social importance.
“When I asked Jesse what he wanted to do with a sense of touch, he said he wanted to be able to hold his wife’s hand,” explains Colgate. “A big part of that is warmth. There’s a big social dimension to this work that is sometimes underappreciated.”
While developing the tactor, Colgate and Peshkin have struggled with a variety of unique challenges. The device must be thin enough to fit in the vest that holds the prosthetic device in place, it must consume as little energy as possible, and it must stay in the appropriate place despite being on a moving body. “At this point it is very experimental. You want to try different capabilities to see what you actually need,” Peshkin says. “Yet even at the experimental level, it’s very tricky engineering.”
A friendly competition
As Kuiken’s research began its clinical phases, the Pentagon also started planning a push for more realistic prosthetic devices. The need for better prosthetic limb systems has become increasingly important as a result of the continuing casualties in Iraq and Afghanistan. Improved body armor and medical treatment have led to higher survival rates: About 90 percent of those injured in Iraq and Afghanistan survive. However, those who survive are increasingly likely to have lost a limb. And while hand and arm amputees make up just 5 percent of civilian amputees, nearly a quarter of the amputees who come home from Iraq have lost an arm or hand.
All four researchers are working on both major projects that have been funded by DARPA to accelerate the pace of prosthetic research — and at times, based on the differing roles and approaches to the project, they even find themselves in competition with one another. Colgate and Peshkin — through their company, Chicago PT — and Richard Weir are developing two different designs for prosthetic arms and hands.
Weir and his colleagues are working on an intrinsic design, meaning that all of the motors and gears are located near the point of use. For example, the prosthetic hand they are designing has 15 miniature motors, with another three in the wrist. Weir’s team works with a variety of corporate and academic partners, including Otto Bock, the leading manufacturer of prosthetic devices. They recently completed their first prototype — a model that brings together a variety of components already in development. As they test that model with patients at RIC, they are busy developing an improved second prototype.
Weir points to one major benefit in their team’s design: It is adaptable for different levels of arm amputation. Because the motors are located near the place of use, they don’t require additional space for a central motor. However, that also limits the amount of space for their equipment. “We have to fit everything into the space of the hand and do it in a form for an average female while providing enough strength for an average male,” says Weir. “It’s very challenging.”
Colgate and Peshkin’s work focuses on developing an arm based on their work in cobotics, a class of robotic devices intended for direct physical collaboration with human operators. In contrast to Weir’s arm, their prosthetic hand is an extrinsic design that runs off of a central motor in the forearm to control the hand through a series of simulated tendons. One major advantage of this technology is that it is inherently flexible, allowing the arm to have a similar amount of give as a human arm.
Like Weir, Colgate and Peshkin also struggle with space and weight issues. Any prosthesis must weigh less than a real arm because it isn’t attached to the shoulder like a real arm. “As an engineer, it feels unfair,” says Peshkin. “In biological systems the actuators are above the body part. Both our actuators and the power supply have to be in the space of the arm itself. Compared with biological systems, you’re always at a disadvantage. It’s really an uphill battle.”
Despite the challenges, the team sees promise in their work.“I think this project may really advance the state of the art,” says Colgate. “Coupled with Todd Kuiken’s research, it has a shot at being really helpful.”
New possibilities for user control
Kuiken’s research in nerve transfers has presented new possibilities for the development of other prosthetic technologies for patients with upper-arm amputations. Other ongoing research between McCormick, Feinberg, and RIC has the potential to provide similar improvements for those who lose their hands, lower arms, or even legs.
Weir is studying one exciting possibility: wireless sensors that could be implanted into the muscle to detect electrical signals and control a prosthetic device. The wireless sensors convey the muscle signals to an external coil around the limb and could be applied to amputees who have lost part of their hand or forearm. Weir hopes that in addition to cutting down the number of wires required for the prosthesis, they will provide a more robust system of reading the body’s electrical signals. In order to better interpret those signals and understand the desired motion of the user, the group is working with Wendy Murray, a new assistant professor of biomedical engineering at McCormick, to conduct fundamental research into the nature of neuromuscular control.
While the research into upper-arm prosthetics is progressing at a rapid pace, Kuiken is also keeping his eye on other opportunities to help amputees. “I have a lifetime of work ahead of me with the arm, but the leg is an exciting area to try as well,” he explains. “There are 10 times as many leg amputees as arm amputees. They’ve just come out with the first motorized legs, and we think we can add steering to them.”
As researchers continue to make revolutionary advances in this field, it’s almost easy to overlook the significance of each step. “I got a call from people working at our company, Chicago PT, telling me that they had tested our tactors and that they had successfully conveyed the sensation of touch to Jesse,” Peshkin says. “I realized that they had done something really special that day. Working with Todd and Jesse gives you the opportunity to do things that never have been done before.”
— Kyle Delaney