A new generation of prostheses that closely mimic, and may one day surpass, the performance of our own limbs could make the ‘bionic man’ a reality. Jon Excell reports.

‘Steve Austin: astronaut. A man barely alive. We can rebuild him. We have the technology. We can make him better than he was. Better… stronger… faster.’

So began one of the cult TV dramas of the 1970s, in which boffins transformed the damaged body of a test pilot into the ‘Six Million Dollar Man’ — a government agent with a right arm as strong as a bulldozer, a top running speed of 60mph, and, of course, a zoom lens for an eye.

In the mid-1970s the technology used to turn Austin from crash victim to bionic man wasn’t much more plausible than Doctor Who’s Tardis, or the transporter on the starship Enterprise. But today, while time travel and teleportation remain firmly in the realm of fiction, advances in engineering, surgery and artificial intelligence are combining to usher in a range of exciting prosthetic devices that closely mimic, and could one day even surpass, the behaviour of human limbs.

And while the latest crop of prostheses won’t turn their recipients into crime-fighting super-heroes, the bionic hands, arms, legs and feet currently under development have a far nobler purpose: to restore mobility and independence to people with lost limbs.

People like Donald Mackillop, a Scotsman who recently picked up a glass with his right hand for the first time in decades. Mackillop, who lost his hand in an industrial accident 30 years ago, has been fitted with i-Limb, a prosthesis that its Edinburgh developer, Touch Bionics, claims is the most advanced bionic hand in the world.

The company, which until a few years ago was part of the NHS, traces its roots to pioneering work on prosthetic limbs for Thalidomide victims in the 1960s. Technology director David Gow believes that the latest product, which will be fully launched later this year, represents a huge breakthrough.

Expected to cost around £8,000, i-Limb comes perhaps closer than any other prosthetic device to mimicking the gentle and economical precision of a human hand. This feat of technical elegance is largely down to the fact that each of the hand’s five digits are individually powered by miniature DC motors and tiny transmission systems.

Thus, while most prosthetic hands are capable only of a fairly rudimentary pincer grip, i-Limb is far more adaptable. ‘We have fingers that articulate and the ability to make more than one grip pattern,’ said Gow. ‘Therefore if we want the patient to have a precision grip they can have a precision grip, if they want a power grip where the fingers do all the gripping we can take the thumb out of the way.’

An important side issue is that the poor geometry of pincer grips means that you have to compensate with far higher gripping forces than we use. Thus, while most human tasks are conducted with forces of just 10-15N, existing prosthetic hands generate forces of around 100N. According to Gow, i-Limb currently falls about halfway between the two. ‘We believe we use the gripping forces appropriate to the task similar to the way the human hand does it,’ he said.

The hand is controlled using electrodes placed close to the skin. These collect electrical impulses from the muscles that are then used to trigger components in the prosthesis. This so-called myoelectric control technology has been used successfully for some time, and is relatively straightforward when applied to pincer-type prosthetic hands with just one motor. But i-Limb has five motors, raising considerable challenges, not only for the developers but also for the patient and the medical staff.

Gow explained that prosthetists will typically work with patients to find optimal muscle signals in two sites on the arm; the muscles that are used to extend or flex the wrist are commonly used. Once these sites have been found, it’s important that the two signals are kept independent of each other, that they are repeatable, and that the electrodes and sensors built into the socket worn by the patient coincide precisely with those sites to ensure good electrical contact.

Sounds tricky. But it gets a lot more complicated with i-Limb, where the performance of the hand and the requirement for multiple electrode sites raises considerable signal processing challenges.

But not all advanced prosthetics require such problematic connection with the body. Icelandic company Ossur is in the early stages of launching the Proprio foot, a £5,500 prosthesis with built-in intelligence.

According to Heidrun Ragnarsdottir, director of R&D at Ossur, the performance of the foot represents a massive improvement over previous systems. Until now the most advanced prosthetic feet available have been passive devices where any ankle movement results from the user putting weight on the prosthesis. The Proprio foot is, by contrast, an active system: accelerometers gather information about the outside world, which is then acted on by a microprocessor-controlled actuator consisting of a worm drive and a stepper motor.

Powered by an external rechargeable battery that helps to minimise the weight of the device and avoids creating an undesirable pendulum effect, the intelligent foot takes just 15 steps to ‘fingerprint’ an individual’s style of gait. Once it has done this, it’s able to use deviations in style to detect different types of terrain and adjust its position accordingly.

When walking on level ground, for instance, the foot will lift the toe during each swing, to prevent it from stubbing the floor. This is a common problem with existing prosthetic feet for which patients compensate by lifting their hip as they walk. Similarly, when walking up a slope the foot will gradually move into the slope angle, making it easier to walk up an incline. And when it detects stairs, the foot will lift the toe so that users can move their centre of mass more easily over the foot, once more avoiding the awkward and potentially damaging compensations that users of traditional prosthetic feet are forced to make.

Ragnarsdottir added that there are relatively few maintenance issues with the Proprio foot. Its ability to adjust to different walking conditions is expected to reduce the requirement for a prosthetist to be on hand to fine-tune the device when someone’s circumstances change. She said that the foot has an expected lifetime of three years, but that this could vary depending on how much use it gets.

Ossur’s efforts are not restricted to the foot. The company has developed a range of impressive prosthetic devices including sockets, shock absorbers, and, intriguingly, the Rheo Knee, a prosthetic knee that exploits the unusual properties of rheological fluid.

The viscosity of this fluid — a suspension of tiny magnetic particles — can be altered by the application of a magnetic field and used to adjust the stiffness of the knee instantaneously. The technology has been used before in both automotive suspension systems and damping systems for earthquake-proof buildings, but according to Ragnarsdottir, this is its first application in prosthetics.


Ossur’s Proprio foot adapts itself to a user’s style of gait

Meanwhile, back in the UK, the creator of the i-Limb hand is turning its attention to the development of a complete arm. Touch Bionics’ David Gow said that the company has already tested the device on the end of an arm developed at the Rehabilitation Institute of Chicago (RIC) but is keen to develop its own all-round expertise. ‘Rather than interface with everybody else’s bits and pieces we now feel that we should develop a complete arm system — the world’s best electronic hand should be at the end of our shoulder, our elbow and our wrist,’ he said.

Ultimately, Gow envisages a plug in-and-play system akin to USB, where elbows, hands, shoulders and wrists can be plugged together and will automatically recognise each other.

To get to this point Touch Bionics is watching with interest pioneering surgical developments on the other side of the Atlantic where last month the RIC introduced the ‘world’s first bionic woman’ — Claudia Mitchell.

Mitchell, who lost her left arm in a motorbike accident three years ago, has been fitted with RIC’s advanced bionic arm that is actually wired into the user’s nervous system. The work builds on a breakthrough made in 2001 when RIC engineers fitted slightly less sophisticated versions of their arm to double amputee Jesse Sullivan.

During a surgical process called targeted muscle reinnervation, nerves in Mitchell’s shoulder, which once went to the amputated arm, were re-routed and connected to healthy muscle in the chest. These nerves have grown into the chest muscle and are used to direct the signals they once sent to the amputated arm to the robotic arm via surface electrodes. Thus, when Mitchell thinks about moving her arm, the bionic arm moves in much the same way as a normal arm.

Claudia Mitchell has been fitted with an advanced bionic arm that is wired into her nervous system

While the idea of wiring prostheses into the nervous system, once unthinkable, is actually happening, engineers are now investigating the even more astonishing possibility of bypassing the nervous system altogether and tapping directly into the brain.

In a development that holds great promise for people completely disabled by conditions like ALS or cerebral palsy, researchers at New York State’s Public Health Department are developing a system that enables patients to send e-mail and communicate using just their brainwaves. The so-called brain computer interface (BCI) uses a non-invasive cap containing electrodes to monitor brain activity and translates these brainwaves into computer commands.

Patients using the system are shown a matrix of icons that flash randomly. When they see the icon that they want to select the resulting spike in the brain’s electrical activity tells the system to select it.

While others have been looking at developing similar implantable systems for the brain, Cambridge Consultants’project manager Mark Manasas — who has been working on the project — said that the non-invasive approach looks more promising. He claimed that while an invasive system will provide a stronger signal, an implant would not only require surgery but would also only be able to pick up signals from a small area of the brain. In contrast, an external cap, while receiving weaker signals, enables sensors to be placed anywhere over the skull, making it easier to tailor the system for individuals’ differing requirements.

One of the big issues with the New York system, said Manasas, results from the need to get good electrical contact for the sensors. He explained that users of the system not only have to have their heads shaved and plastered with electrolytic gel, but they must also wear a metal cap with tiny spikes that penetrate the skin.

‘Once you find the good locations for the electrodes you don’t want to vary those,’ said Manasas. ‘You’re tuned to certain locations, but if you put these tiny little spikes into your scalp every day to get good electrical conduction you’re ultimately going to get scarring and the tissue’s going to get scabby — it’s a mundane problem but it’s one of the bigger issues in non-invasive BCI.’

While the system’s most promising current application is enabling ‘locked-in’ people to communicate with the outside world, Manasas said that it could easily be developed to help patients to perform mundane tasks such as turning lights on and off. However, while the technique has been used to control a robot arm, he said that the challenges of using it actually to move prosthetic limbs would be immense.

‘It’s not really hard to tell an arm to move in this direction or that direction, what’s really hard is to dwell. It’s easy to get a binary signal, but to say “move this far and then stay there” is much more difficult.’

Instead, Manasas believes the next big thing in prosthetics will be the development of implants that connect directly to the nerve endings, and wirelessly relay signals to the limb. This will call for precise antennae that are able to direct the signals from these implants at a point in the prosthesis. ‘The ultimate goal is to get people doing something as critical as driving a car. If a prosthetic limb was to perform the wrong movement it could be catastrophic.’

There are clearly some huge challenges to overcome. But with so much advanced prosthetics research already bearing fruit, it seems likely that the increasing interface between machine and nervous system will constitute one of the major developments in the medical devices industry for years to come.

As a result, in the not so distant future it may be difficult to tell where the human body ends and the prosthesis begins.