Humans are incredibly adaptable. A person can lose a hand or a leg and learn how to do most of the same things they could do before, from mundane daily tasks to impressive athletic feats. No one would argue that life is easier when you’ve lost a limb  though, which is why we’ve been making, using, and improving prosthetic limbs for thousands of years.

The earliest prosthetics were rudimentary limbs made of bundled fibers, wood, or metal. The Roman General Marcus Sergius was said to have worn an iron hand circa 200 BC. These limbs were attached with straps and harnesses, and didn’t offer much functionality—a wooden leg might offer balance and slow walking with a cane, but an iron hand or prosthetic arm was mostly just cosmetic.

Prosthetic limb technology advanced slowly, but the development of better attachment systems, lighter materials, and more effective control mechanisms improved life for those who had lost limbs or were born without them. Major advancements were often driven by sudden influxes of wounded soldiers who lost limbs in the world’s most cataclysmic wars. For instance, the number of amputees coming home from Civil War battlefields drove researchers to create several new types of artificial limbs, including the Hanger Limb and William Selpho’s prosthetic arm.

Selpho’s design is particularly significant because it included one of the early control mechanisms, offering something beyond a simple static hook. A strap running to the opposite shoulder could be pulled taught by shrugging or otherwise moving the shoulder, activating a mechanism in the prosthetic hand that caused the fingers to close and open. While prosthetic technology seemed to creep forward for the last few hundred years, it’s clear that important work was being done, and the advances helped improve lives.

Advances in the last decade, however, have moved with astonishing speed. New materials like titanium and carbon fiber make today’s prosthetics lighter, which brings a host of add-on benefits like greater freedom of movement, reduced user fatigue, and a wider variety of attachment options. Cutting edge control mechanisms are far beyond anything Selpho could have imagined, incorporating muscle impulses or even brain-computer interfaces.

One of the exciting things about prosthetic development is that it incorporates so many scientific disciplines. Medical science works with the prosthetic wearer’s physiology and the underlying tissues of a residual limb. Materials science and engineering play a major role in development. Practical nuts and bolts mechanical know-how is important too. Computer science has taken on a vital role with the development of more advanced control systems. While this may have once created barriers to smooth, efficient development, researchers and people involved with open source prosthetic projects can take advantage of the Internet to share information and build off of each other’s work.

Materials

Advanced plastics have improved the cosmetic appearance of prostheses for those who desire a natural appearance. Shells and covers can be carefully matched to the wearer’s skin tone, and expert painting can capture the look of real skin.

The biggest breakthrough in prosthetic materials has been carbon fiber. Carbon fiber is significantly stronger than steel while weighing much less. It’s made of fine carbon filaments, which are woven into larger fibers, made into fabric and laminated with epoxy resin into rigid shapes. Its properties allow it to be constructed into shapes that bend and rebound, acting like springs. The best-known example is the Flex-Foot Cheetah, a carbon fiber prosthetic leg that looks like a curved “blade.” Many para-athletes use these legs, which allow them to run with a natural gait at high speeds.

The lightness of carbon fiber allows prosthetic limbs to be attached with suction or with ratcheting devices that are much more comfortable and easy to use than earlier strap systems. A split-toe carbon fiber foot flexes naturally, increasing the wearer’s agility and the naturalness of his or her gait.

Control Methods

No aspect of prosthetic science has developed as much, or become as complex, as control methods. Lower limb prosthetics are somewhat simpler because the wearer can control it without any kind of command-response mechanism. A good example of this is the Rheo Knee. A human knee is not a simple hinge that flops along as you swing your legs. Through complex muscular controls you may not be aware exist, the stiffness and angle of your knee varies depending on your walking speed, the surface you’re traversing, and even where you are in your stride. The Rheo Knee uses microcomputers to measure the load and angle experienced by the knee. It then adjusts the stiffness of the joint by altering the viscosity of a magnetic fluid. Another common prosthetic, the C-Leg, also provides computer controlled dynamic stiffness adjustments. The result is not only a more natural gait, but the wearer has an easier time walking because the leg properly transfers momentum during the various phases of a stride.

The next step in control mechanisms involves myoelectrics. A myoelectric system gives the wearer direct control over some of the prosthesis’ movements, the degree of which largely depends on the tissue and muscle left in the residual limb. Those muscles still receive signals from the brain when a command to move the limb is issued, even though the limb is no longer there. Normally, those nerve signals dead-end, but myoelectric sensors pick up those signals and interpret them into controlled movements of a prosthetic limb. In some cases, nerves are surgically rerouted to send signals to different existing muscles.

To use a myoelectric limb, the wearer has to undergo some training, and the device must be adjusted until they “learn” how to work together. By tensing specific muscles, the wearer can induce specific movements in the prosthetic. Different intensities of the same muscle can create different movements. These commands can be built up into very complex movements, including sequential finger movements or large shoulder motions. Even in cases where nerve damage or a lack of sufficient remaining muscle in the residual limb would prevent myoelectric operation, muscles in the chest and back can be used instead.

Myoelectric arms and hands are already on the market. The bebionic3 hand and a number of elbow, wrist and hand mechanisms from Ottobock are just a few examples. Myoelectric legs provide even greater freedom of movement compared to a dynamic leg like the Rheo Knee, because the wearer can activate motors to lift the leg up stairs or even adjust it while sitting without manually moving the leg by hand.

The concept of myoelectrics ultimately leads to brain-computer interfaces. If the control system for our muscles is really just a network of signals and relays, why not cut out the relays and go directly to the command center? This may be necessary because a spinal injury prevents any nerve signals from moving through the body, making myoelectric control impossible.

Whenever something happens in your brain, whether you’re reflecting on a memory or deciding to make a fist, electric signals are generated as the ions in each neuron create a difference in electrical potential. These signals can be read either by electrodes inserted surgically into the brain or by electrodes attached to the scalp. There are problems with both methods. The signals are quite tiny, and your skull is a pretty good insulator, so scalp electrodes can pick up crosstalk or just have a hard time getting a reading.

Implanted electrodes get better signals, but of course they require invasive surgery. They need to be placed in the correct place within the brain to read the signals necessary to control a device. Unfortunately your brain is not an unchanging thing—over time it grows and shifts so that the implanted electrodes slowly migrate away from where they need to be.

While controlling prosthesis with the mind is an enticing idea, and one with enormous potential, there are a lot of difficulties with brain-computer interfaces. We’re probably a few decades away from the kind of smoothly responsive, accurate and effective control you might see in a science-fiction story.

Sensory Feedback

Controlling a prosthetic limb has been a one-way street for many years. The wearer issues a command, the limb receives the signal and performs the task. Only recently has progress been made to send signals the other way, giving the prosthesis a sense of touch. This would allow the wearer to experience someone holding his or her hand, or more easily accomplish tasks like reaching into a bag to grab an apple instead of a pencil. It’s also very important for fine tasks like holding a delicate object. Research in this area is not as well developed as controlling a limb, but DARPA’s FINE (flat interface nerve electrode) already shows a great deal of promise.

Open Source

There are three primary reasons for the rapid advance of prosthetic technology in recent years. Increased availability and expertise in working with carbon fiber, along with improvements in carbon fiber manufacturing that make it easier and less expensive to work with, have had a huge impact. Another major factor is DARPA (Defense Advanced Research Projects Agency) initiated their Revolutionizing Prosthetics program in 2006, spurred in large part by the large number of soldiers returning from Middle Eastern conflicts with amputation injuries. Research backed by DARPA support and funding produced several commercially viable prosthetic systems, and major advances in brain control.

The third factor might be the most exciting—open source prosthetics. Prosthetic wearers have always tinkered with and improved on their limbs, of course, but there was never a method of easily sharing what they’d learned or made, and no framework for working on projects together. The open source concept, in which designs are shared freely with the public so they can be used, adapted, and improved upon, has pushed prosthetic design in new directions and helped push down prices.

A quick look around the various community projects ongoing at The Open Prosthetics Project shows an astonishing array of development. One group found that plastic zip ties make excellent artificial tendons. Another is using Lego building blocks to prototype hand designs. A project to improve attachment methods (called “suspensions”) hit upon the idea of the Chinese finger puzzle: a sleeve open at each end that grabs onto a finger (or, for instance, an arm stump) with greater strength the harder it’s pulled due to the alignment of fibers within the sleeve. Their early experiments have been promising. Yet another group found a few versions of a classic “split hook” artificial hand that is no longer produced. They hope to reverse engineer it and make new, cheaper versions with modern materials.

If the progress of prosthetics follows a predictable arc, we’ll see artificial limbs get lighter and stronger in the coming years, as control methods become more reliable and sensory feedback more lifelike. But the most important advance might be making quality prosthetics less expensive. An advanced prosthetic leg or arm can cost tens of thousands of dollars, and few insurance policies will cover the price.

Open source holds a lot of promise. However, advanced prosthetics will remain out of financial reach for most of the people with amputation injuries or congenital limb problems. There are several charity and outreach organizations, including the Limbs for Life Foundation and the Amputee Coalition. The Open Prosthetics Project even has a wiki page with information on getting financial assistance to buy a prosthetic limb.

In Kansas City, a young boy with congenital deformity of his right hand couldn’t afford a commercially produced prosthetic. A local teen used shared files by the open source community to create a hand using the local library’s 3D printer, then assembled the hand for the young boy to use. The hand’s designers even adapted the design to fit the boy. It’s certainly a heartwarming story, but it’s also a great example of the benefits of open source prosthetics.

Share this page on: