Building natural movement into prosthetics

Prosthetists can fashion artificial limbs that look remarkably like the real thing. Getting them to perform like the real thing was the goal of a National Science Foundation-funded project conducted by:

  • Dinal Andreasen, GTRI senior research scientist,
  • Ravi Bellamkonda, professor of biomedical engineering, and
  • Isaac Clements, Graduate Student of the Coulter Department of Biomedical Engineering at Georgia Tech and Emory University.

The research team worked on a peripheral nerve interface to capture the nerve signals between the brain and a lost limb, according to Andreasen. Those signals could be used to operate a prosthesis.

The device consists of a cap placed over the stump at the time of amputation. Inside the cap is a scaffold of parallel polymer nanofibers, each about 500 nanometers across, that acts as a kind of trellis. "The idea is to encourage regeneration of the neurons into the scaffold, which also contain electrodes that the neurons grow over and come into contact with," Andreasen said.

Nerve signals going to the limb are picked up by the flat gold electrodes. They are sent to the outside of the cap, where special receivers enable researchers to monitor signals transmitted from individual nerve fibers.

Taking Advantage of Natural Growth

Most current approaches to this problem either cannot discriminate between many fibers firing simultaneously, or do not support much growth near them. The GTRI-Georgia Tech team’s approach took advantage of natural growth processes and at the same time, could provide more precise control of artificial limbs.

"A nerve is like a cable that consists of thousands of individual fibers called axons," Bellamkonda explained. "Each axon is hooked up to a different part of an organ or muscle unit or other part of the body." The nanofibers in the cap control where the fibers grow, "so presumably you can listen to larger numbers of individual fibers and therefore have a finer sense of the intent of that particular signal," he added.

Once the individual nerve impulses could be isolated and their signals interpreted, they could be converted into multi-channel electrical signals. The signals would then be used to control a prosthetic device.

Since nerves carry both motor and sensory information, the technique could also lead to prosthetics that provide sensory input.

Andreasen noted that moving one's fingers, for example, involves sending the command to move and receiving feedback about where the fingers are, whether or not they have come into contact with an object, whether that object is hot or cold, and so forth.

"A major part of the research involves sending stimulation signals back into the peripheral nerve to give you the feeling that your fingers are really moving," he said. "Working with information flowing both ways through the peripheral nerves is an interesting and challenging part of the project."

Although the ideal scenario would be to attach the interface at the time of amputation, it may not be a requirement for the device to work, Andreasen noted. "The nerves remain intact many years post-amputation, and we are confident we can induce them to grow into our regenerative electrodes."

How long such a system will be stable and how much control an amputee can attain from such a system remain open questions, he added.