O&P researchers are hard at work to meet the demands of today’s amputees, who insist on prosthetic technology that fits with their lives. Whether running errands or running marathons, amputees need their prostheses to provide support — and springiness — to keep them on their feet.
In humans and other animals, stiffness and damping in the joints are necessary functions of locomotion.
Humans are basically an inverted pendulum, with their back legs swinging over their outstretched front legs, explained Roy Kornbluh, senior research engineer in the engineering and systems division of SRI International in Menlo Park, Calif. The natural frequency associated with gait is determined by the length of their legs and their mass, among other factors.
At various steps in the gait cycle, the tendons and muscles assume the responsibility for storing the energy and delivering it at the right times, which makes it possible to walk without exerting a great deal of energy. If that cycle is interrupted — by amputating a person’s leg — that person’s natural gait suffers.
Tunable prostheses are devices that can be adjusted based on the wearer’s current conditions, including terrain, weight and speed. They allow the amputee to control when stored energy is delivered, just like with a human leg, giving the amputee a more natural gait.
Current tunable technology focuses on controllable damping, which dissipates the amputee’s energy in a controlled manner, much like a brake dissipates a car’s energy. Prostheses such as Ossur’s Rheo Knee and OttoBock Health Care’s C-Leg give amputees the ability to vary the damping of the prosthesis through computer-controlled, sensor-driven modulation of the knee.
“That is one step in the right direction to making a more natural and efficient gait, but that is just how it absorbs energy. That is an easier problem,” Kornbluh said. “The other part of the equation is the springiness, and that is harder to implement mechanically.”
Choosing correct stiffness
The issues in question are, first, how much stiffness is necessary for the amputee at the current time? Second, when will that stiffness, or springiness, be most useful? In order to fully replicate natural gait, a prosthesis must not only provide the proper stiffness, but do it at the right time.
“Storing the energy is basically loading a spring,” Kornbluh said. “If you squeeze a spring, you are storing the energy in it. As long as you hold that spring, it has that energy stored and then boom — you could release it any time you want.”
Now think of a sprinter: He would require a prosthesis with a stiff ankle joint to speed around a track. After finishing the race, however, the sprinter probably would switch to a different prosthesis with a softer ankle joint that allows for enough flexing, even though the higher forces he experienced when running are gone.
Likewise, when he walks off the track, he might not need the immediate spring he had during the race. Kornbluh said that the sprinter might want to absorb the energy as he steps — hold it — and then have the prosthesis spring back on a bit of a delay, depending on his walking gait.
With a tunable prosthesis, the sprinter would be able to do both.
One of the frontrunners in this technology’s advancement is Hugh Herr, PhD, associate professor at Massachusetts Institute of Technology (MIT) in Cambridge and director of the biomechatronics group at the MIT Media Lab. Along with an extensive background in engineering and physics, Herr also is a bilateral transtibial amputee, and so sees the need for this development from all sides.
His research was integral to the creation of the Variable-Damper Knee Prosthesis, which Ossur commercialized as the Rheo Knee in 2005. Herr — along with mechanical design from Gill Pratt, PhD, former director of the MIT Leg Laboratory — developed a prosthesis that uses magnetorheological fluid to automatically adapt knee damping to the amputee’s gait by sensing knee force, torque and position, according to the MIT Media Lab Web site.
This technology represents a breakthrough for the O&P profession, but is only the beginning for tunable prostheses. For instance, the Rheo Knee allows amputees to tune damping, but not springiness.
“Because of that, when amputees walk in the Rheo Knee, they walk more slowly and their metabolic rate is higher compared to people with intact limbs,” Herr said.
“It is easier to control damping because brakes and shock absorbers are easier to control than springs,” he said.
Benefit for amputees
The main benefit for amputees is the ability to conserve energy. Because they do not have equal and efficient energy absorption and delivery, amputees must exert more energy than able-bodied people to walk with conventional unpowered prostheses, Kornbluh said.
When using this new technology, the prosthetic user would be able to store the energy he has created in the loading part of stance phase until he needed it, instead of allowing the spring to expand automatically. This ability would match natural human locomotion, where the leg can respond to any number of disturbances the person introduces.
“When changing speed and terrain, the intact human ankle varies stiffness and damping continuously,” Herr said. “When a person picks up something heavy, [his] joints automatically respond. A person with intact limbs can walk faster, walk over different terrain surfaces, put on a different shoe, and [his] leg joints respond by varying stiffness and damping.”
Researchers and manufacturers are beginning to realize the value of these abilities in prosthetic design.
Thus far, several means of packaging this technology have been created.
One option for implementing controlled springiness is with an artificial muscle, a device that expands and contracts with stimulation, much like a natural muscle. Kornbluh said that, when controlling only stiffness or damping, only a small amount of power from an outside source — an artificial muscle — is necessary.
The type of artificial muscles that Kornbluh and his team at SRI International have developed consist of soft, rubbery polymers that change length when stimulated, as by electricity. These are a viable option for prostheses because they offer the force, stroke, springiness and light weight of natural muscle. Pneumatic or hydraulic artificial muscles also could offer muscle-like behavior — with an extra motor or compressor — but the muscle itself still is lightweight. (For more information about artificial muscles, see “Strength in Innovation,” in the July 1, 2007 issue of O&P Business News.)
Electrical artificial muscles large enough for prostheses are still a few years away. In the meantime, Kornbluh and his team at SRI International have been working to develop “electrolaminate” technology, which varies stiffness. Kornbluh explained the technology.
Layers of relatively rigid material are interspersed with layers of softer, rubbery material. Using electrostatic clamping, Kornbluh and his team decide whether or not the rigid layers can slide freely relative to the softer layers. This result makes it look like the stiffness either of the rigid layers or of the softer layers, when electrically activated by clamping together those layers, or unactivated and allowing free sliding. That result also could consist of any level of stiffness in between the most rigid or the softest, depending on how many layers are clamped.
“The advantage to electrolaminate technology is it could be part of the structure — it is not an extra device,” Kornbluh said.
He offers the example of thin carbon fiber prosthetic feet, which have a layered composite structure. A structure such as this, when combined with electrolaminate capability, would allow stiffness tunability with minimal effort from the user, and only a small amount of power.
“Something the size of a 9-volt battery could probably last for many days,” he said.
Ideally, this stiffness tunability would be dialed in by amputees based on their current needs, sensor-based, or some combination of the two.
Variable stiffness and damping
The prosthetics world need not wait long for new tunable stiffness and damping technologies to emerge from the laboratory. Herr and his team at MIT’s Media Lab currently are working on two new systems: a powered ankle foot prosthesis and an orthosis. These motorized devices allow for continuous variable stiffness and damping, depending on the amputee’s gait cycle, speed of walking and the terrain.
Both systems feature “fairly sophisticated control algorithms where a motor system actually controls forces that it applies,” Herr said. “Then because of that, you can simulate spring-like behavior or damper-like behavior using the motor system.”
The underlying advantage to these systems is that the software can output any force within limits. When amputees move their joints, they feel something like a spring across the joints of that stiffness, Herr said. Such systems require a battery, but it is worth the reward of a high degree of variability.
In addition, given the lightweight components, the system fits inside a sleek design.
“We are able to package the system within a human-like envelope,” he said. “It is not bulky. It is cosmetic.”
PowerFoot, the powered ankle foot prosthesis, will be publicly available in 2009, from iWalk Inc. The system currently is being tested with unilateral amputees, as well as with Herr, the only bilateral amputee to test the device as of press time.
Currently, the active ankle-foot orthosis is being commercialized to assist patients with footdrop.
The future of tunable prostheses is coming soon, according to Kornbluh. With much of this technology already demonstrated in various arenas, it simply requires more sophisticated mechanics. SRI’s electrolaminates, for example, should be available within the next few years.
“I think the issue is that, unless you find something like a DARPA prosthetics program, there are not a lot of research dollars for radical changes in the O&P community,” he said. “That is why it almost has to come from work in robotics or another area.”
With some of the profession’s top researchers working on this technology, it is only a matter of time before it reaches mainstream use.
A prosthesis must be able to stiffen, dampen and thrust in order to replicate human leg mechanics, Herr told O&P Business News. If the prosthesis can accomplish all three, within the confines of a reasonable weight, it could solve major clinical problems, such as walking speed, metabolic cost and stability.
“In the next generation of systems, we certainly will improve stability compared to conventional prostheses,” he said. “But to completely restore stability is, in my view, the final frontier, and may require a better neural interface so that amputees can actually feel the bottom of their feet.”
Nerve innervation may be the road to this, though proprioception is still far in the distance.
Herr projects that the field will see robotic prostheses capable of varying stiffness, damping and motive output in order to solve pathology in walking speed and metabolism, and thereby vastly improve amputees’ ability.
“There are a lot of prosthetists and orthotists who do not like the idea of robotic limbs. They believe such devices will always be too heavy, too large, and they will break frequently,” he said. “Fairly complicated systems can be made robust. That is where the field is going to go.”
For more information:
Stephanie Z. Pavlou is a staff writer for O&P Business News.