A brave new world

In a story that should belong to the annals of science fiction, an international team of research scientists are developing a prosthetic arm and hand that has sensation and can be moved by thought, imitating many of the movements and qualities of the natural limb. Callan Emery reports.

Scientists in the United States and Europe are collaborating on the development of a thought-controlled prosthetic arm that is decidedly futuristic in the way it works. It can move through multiple degrees of freedom and provide sensory feedback for touch and temperature.

Watching a video of the bionic arm being thought-controlled by one of the trial volunteers, you’d be easily forgiven for thinking this scenario is straight out of a science fiction movie. It is a strangely eerie portrayal of the first man-robot combination as the person uses the bionic arm, attached at the shoulder, to iron a pair of pants on an ironing board or cut vegetables to add to a salad. The arm moves in that peculiarly robotic style.

The initial research, which began around 2001, now has more than 12 volunteers trialling different versions of the bionic arm at the Rehabilitation Institute of Chicago in the United States.

This futuristic prosthesis is being developed by a team of international research scientists -- biomedical, electrical, mechanical engineers, neuroscientists, physiatrists, surgeons, prosthetists and therapists – in a virtual collaboration known as the Revolutionizing Prosthetics 2009 (RP 2009) team.

The team is being led by Johns Hopkins University Applied Physics Laboratory (APL) in Laurel, Maryland, US and is funded by DARPA – the rather darkly apocalyptic sounding acronym for the equally portentous US defense department’s Defense Advanced Research Projects Agency.

The US$55-million four-year project has an ambitious directive – develop a next-generation mechanical arm and hand that will look, feel, perform and be controlled like a natural limb. The prosthesis should be able to pick up a cup of coffee, button a shirt, put a bank card into an ATM, feel the warmth of a partner’s hand, enable more natural walking by swinging freely at the person’s side and provide propioception, the ability to sense where your limbs are when your eyes are closed.

Thought control

The thought-controlled bionic arm is the brainchild of Todd Kuiken, MD, PhD, director of the Neural Engineering Center for Bionic Medicine at the Rehabilitation Institute of Chicago (RIC).

He fitted the first bionic arm – Proto 1 – to trial volunteer Jessie Sullivan in 2002 at RIC. The arm was publicly unveiled in 2005. Since then the project has come a long way. In August last year RP 2009 unveiled and demonstrated Proto 2 at the DARPA Tech 2007 conference in Anaheim, California.

So how does this bionic arm work? To provide the thought-controlled movement of the bionic arm the brachial plexus nerves, located in the amputee’s shoulder and which, prior to amputation, extended into the arm, are re-routed and connected to muscle in the pectoral area of the chest. This surgical process – pioneered by Dr Kuiken and first demonstrated in 2001 – is called targeted muscle reinnervation (TMR).

TMR enables the re-routed nerves to grow into the appropriated muscle and direct the signals they once sent to the amputated arm instead to these muscles. Electrodes placed on these muscles at the sites of reinnervation pick up the nerve signals as the muscle contracts and convey them as an electromyogram to microprocessors on the bionic arm. The microprocessors decode the electromyogram inpulse signal patterns and, in turn, instruct various motors driving the joints of the arm when and how much to move.

When the patient thinks about moving his or her arm, the action is carried out as voluntarily as it would be in a natural arm, allowing a relatively smooth and controlled movement of the prosthesis.

Muscle-electrode, or myoelectric, powered prostheses have been around for some time – such as Touch Bionics’ relatively sophisticated and recently commercialised i-Limb Hand – but none use the muscle reinnervation technique and most are limited to two simple movements. The i-Limb Hand, launched to market late 2007, can only open and close, for example, although the prosthetic hand has five individually powered digits, which are intelligently computer-controlled with a sensory force-feedback system. (See The Hand 32).

Muscle reinnervation

Blair Lock, a biomedical engineer at RIC working on the bionic arm project, explained that the researchers reinnervate the four major nerves of the brachial plexus to the chest muscles. The brachial plexus carries all of the motor information to the arm.

Lock added that surgical techniques continue to be refined to utilise specific motor fascicles of these nerves to provide more detailed signalling patterns and said there was a lot of research into “pattern recognition” taking place currently.

“Interestingly,” Dr Kuiken noted in a research paper, “We have found that muscle reinnervation still works if surgery is carried out up to 9-15 months after the amputation.” This shows that these neural pathways remain intact, even when they have not been used for awhile.

Dr Kuiken also noted in a research paper specifically about the subject, that volunteers who have had TMR have noticed that there is a redirection of sensation to the chest skin of the amputees.

TMT “allows some sensory nerves from the amputated limb to reinnervate overlying chest skin. When this reinnervated skin is touched, the amputees perceive that they are being touched on their missing limb.”

In one individual it even allowed for the cutaneous expression of proprioception, he pointed out, adding that this may allow for “useful sensory feedback from prosthetic devices and provides insight into the mechanisms of peripheral neural plasticity and regeneration in humans”.

Lock explained that the electrical signals of the nerves are very small, “thus the advantage of targeted motor reinnervation.

Lock said that the “reinnervated muscle acts as a biological amplifier of the nerve signals. The electrical signals resulting from the muscle contractions (representative of the nerve signals) are picked up on the skin surface by standard electrode technology.”

However, the standard electrode technology has been taken a step further. RIC research scientist Richard F. Weir, PhD, has developed Injectable MyoElectric Sensor (IMES) devices for Proto 2. The IMES are very small injectable or surgically implantable sensors that are used to measure muscle activity at the source, providing greater signal isolation and better performance compared to the surface electrodes on the skin.

Lock explained that the researchers are generally using 10-12 electrodes to control 10- 14 movements, although they have used up to 128 electrodes specifically to “investigate information available from a high spatial density of electromyogram recordings”.

25 degrees of freedom

Compared to Proto 1’s eight different kinds of movement – a level of control far beyond the current state of the art for prosthetic limbs – Proto 2 can move through 25 degrees of freedom, including movement of the shoulder, elbow, wrist, hand and fingers. It has the strength and speed of movement approaching the capabilities of the human arm, combined with more than 80 individual sensory elements for feedback of touch, temperature, and limb position.

Lock said the arms are “fitted in the same manner that a conventional prosthesis would be. That is, using standard- of-care prosthetic socket systems. There is research ongoing, however, to improve these systems”.

“Each arm being tested is made from a different material. The most advanced arms are being designed to be very light weight, durable and very strong in all degrees of freedom,” he said.

Dr Kuiken said: “The results we are achieving in this highly collaborative project are very exciting.” He expressed his confidence that these discoveries would bring more natural control of prostheses, better artificial limbs and make a difference in the lives of amputees worldwide.


What challenges remain and what does the future hold? A recent article in Popular Mechanics – which honoured the team with the 2007 Popular Mechanics Breakthrough Award – said that the weight of the arm was one of the key challenges. The research team has a target weight of 3.2 kg, while the full Proto 2 arm weighs 4.3 kg. The other hurdle, the article noted, is power. To remain powered, Proto 2 has to be connected to a battery pack which requires a battery belt pack to be worn all the time. And the researchers have not yet met DARPA’s requirement for 18 hours per charge.

Proprioception remains another major challenge. Currently amputees have to watch their arm carefully while controlling it. Perhaps the discovery of the redirection of cutaneous sensation – and in one individual, proprioception - will provide the answer.

In future, research should also provide the opportunity to track and measure how brain mapping changes in patients who have received this procedure. According to RIC the researchers hope to demonstrate neuroplasticity, which allows amputees to adapt to the physical changes made by the surgical reeinnervation technique. Locked noted that subjects become more proficient with use of their prosthesis with practice.

In 2006, the first-ever motorised prosthetic leg was introduced. According to Lock, RIC “wants to develop the steering mechanism for these prosthetics”. He says Dr Kuiken is actively exploring research in this direction.

 Date of upload: 3rd April 2008

                                               Copyright © 2008 All Rights Reserved.