Self-learning Prosthesis Offers More Natural, Fluid Movement
July 02, 2018
Scientists at Imperial College London, England, and the University of Göttingen, Germany, have used machine learning to improve the performance of a prosthetic hand. The device uses a human-machine interface that interprets the patient's intentions and sends commands to the artificial limb. It contains eight electrodes that pick up weak electrical signals from the patient's residual limb before amplifying them and sending them to a computer in the prosthesis. The computer then runs the machine-learning algorithm to interpret the signals before commanding the hand's motors to move in the way the patient wants.
"The new bionic hand is not only more natural, but it is also superior in terms of functionality in daily tasks than what's currently available to patients," said Dario Farina, MScEE, PhD, a professor at Imperial College London's department of bioengineering. "When designing bionic limbs, our main goal is to let patients control them as naturally as though they were their biological limbs. This new technology takes us a step closer to achieving this."
After testing the prototype on five people with amputations, the researchers found that new machine learning-based control was better at providing natural, fluid movements than available technology. The participants found they were able to easily rotate the wrist and open the hand either simultaneously or separately. They also found the movements more natural than the conventional bionic limbs they were used to. In addition to types of function, the participants could also control the speed of individual movement independently of other movements.
Prior to use, the participants underwent training so the algorithm could learn how to interpret their unique electronic signals. Farina hopes to eliminate the need for this step in future prototypes, without sacrificing personalization .
The researchers are currently working to gain greater control over the hand, including the ability to move individual fingers, and eliminate the need for electrodes by transferring signals wirelessly within the patient's body. They hope to have the prosthesis on the market within three years.
Farina is the senior author of the paper describing the results, which was published June 20 in Science Robotics.
To read more about the research that led to the study, visit Nerve Signals Used to Control Prosthetic Arm.
Editor's note: This story was adapted from materials provided by Imperial College of London.