The quest for a smart prosthetic

Disabled U.S. vets could soon be wearing artificial limbs that work like they're real.

Bionic arms and legs that feel heat and pressure, and move with ease, may be available within a few years for U.S. troops who return without limbs from Afghanistan and other battle zones. The prosthetics would perform with Luke Skywalker precision, and be operated by users’ minds in the same way they used their limbs before, and all by way of two-way fibre-optic sensors.

The linchpin for this revolutionary development is a just-launched two-year $5.6-million research project led by Southern Methodist University (SMU) in Dallas and funded by the Defense Advanced Research Projects Agency (DARPA). While the technology offers endless real-life applications, it was the advanced prosthetics potential that led DARPA to fund it. The agency hopes the creation of a “smart prosthetic” will improve the lives of military amputees.

The project’s goal is to develop a sensor that can carry nerve signals through synthetic channels, and create an interface for both sensing and stimulation. “When a nerve spike happens, it’s actually a flow of ions in and out of the nerve body. And that flow of ions is moving charges, so we can sense a very small electric field close to where that happens,” says Marc Christensen, chair of the department of electrical engineering at SMU. To achieve its goal, the project will proceed in two parts: developing the sensor, and testing its biological applications.

Christensen and collaborator Volkan Otugen have built prototype sensors, but they’ve never embedded one in a person. The sensor uses an optical fibre and a light coupled to a tiny silicone sphere. When a light wave travels around the sphere, it creates resonance peaks of brightness that can be sensed when the light re-enters the fibre. If connected to a nerve, the light flow will be altered by the firing of a nerve impulse. Any such change can be detected when the signal travels back into the fibre, which is just 40 microns in diameter — less than half the diameter of a human hair.

Harnessing such quick, hypersensitive two-way optical communication is what will give a prosthetic realistic range of movement. “If you were to sit and wiggle your fingers like you were playing the piano, there are many degrees of freedom that the limited number of electrical interfaces can’t provide today,” says Christensen. Some prosthetics have been produced with electrical interfaces that have hundreds of sites, but they don’t move naturally and typically fail within two to six months. “We need a solution that will work at least 10 years,” says Christensen.

The project’s second phase will focus on combining the sensors with the body. A variety of specialists will work together to design a peripheral nerve cuff to make this possible.

To picture it, think of a nerve as a telephone cable with a bundle of wires inside. In the body, the cable is biological, and the signals are electrochemical. The goal here is to allow the synthetic and the biological to connect and interact. As such, an elastic cuff gently presses fibres toward the outside of that nerve bundle. Inside the cuff, the silicon spheres will stimulate the nerves and carry the detection sensors. “If the person wants to reach out and grab a coffee cup, their thinking about that will cause signals to travel down the peripheral nerve” says Christensen. Then, they’ll be picked up by the sensors just as they would be by the nerves in the hands and arms in a normal body.

The technology would also allow for the mapping of nerve signals so that when a person thinks “Reach out for the coffee cup,” the robotic arm does exactly that. “When they pick up the cup, they may feel the Styrofoam give a little bit or they might feel the heat of the coffee,” says Christensen.

Whatever the short-term goals, it’s easy to see how this work has the potential to effect huge change in neurology and optics. The sensors could help patch a spinal-cord injury and create a better cochlear implant, for example. “This is akin to a rocket launch or developing a transistor,” says Christensen.

He is also excited about other applications. “Prosthetics is obvious, and that’s the one we held up, but this is really a general interface between the brain and computing technology,” he says. “We’re excited about another sensing modality into the brain, another insight into neurological disorders like Alzheimer’s, Parkinson’s and epileptic seizures. They could put a sensor into this area and see what’s really happening.”