Researchers from Ecole polytechnique fédérale de Lausanne (EPFL) research institute in Switzerland have developed a prosthetic hand that uses artificial intelligence (AI) to adapt to a user’s finger movement.
Primarily for amputees, the new technology combines neuroengineering and robotics to enable better grasping and manipulation.
Meanwhile, robotics allows the hand to grasp objects and keep them in contact.
The technology is designed to decipher user intentions and translate them into finger movement of the prosthetic. Users will need to perform a series of hand movements to train the algorithm, which is powered by machine learning.
Sensors will be located on the amputee’s stump to identify muscular activity. The algorithm uses this information to learn hand movements corresponding to particular muscular activity patterns.
Insights into the user’s intended finger movements help to regulate each of the prosthetic’s fingers, said the researchers.
Furthermore, researchers designed the algorithm to initiate robotic automation when the user tries to grasp an object. The algorithm instructs the prosthetic to close its fingers when an object comes into contact with sensors placed on the prosthetic hand’s surface.
The team tested the new neuroprosthetic technology on three amputees and seven healthy volunteers. The Nature Machine Intelligence journal contains published data from the study.
Testing of the algorithm is currently being carried out on a robot. Further research will take place before the commercial availability of the technology.
EPFL Translational Neuroengineering Bertarelli Foundation chair Silvestro Micera said: “Our shared approach to control robotic hands could be used in several neuroprosthetic applications, such as bionic hand prostheses and brain-to-machine interfaces, increasing the clinical impact and usability of these devices.”