Higher School of Economics has created a model for decoding prosthetic finger movementsResearchers have developed a model that determines the movements of prosthetic fingers based on signals from forearm muscles. The press service of the National Research University Higher School of Economics reported about the development "Hitek".
Researchers from the National Research University Higher School of Economics have developed a model for restoring finger movements based on electromyographic signals of forearm muscles in people with disabilities. The accuracy of the model applied to the data of a subject with congenital hand absence was 50%, and 71% for an adult amputation survivor.
The scientists used a wireless 8-channel bracelet and a virtual reality helmet to record EMG activity (electromyography) in healthy subjects. During the experiment, participants worked in a virtual environment with the ability to capture and store the coordinates of individual fingers.
Participants performed symmetrical movements with both hands to complete tasks in the VR environment, and the system recorded finger coordinates and electrical activity in the forearm muscles. The researchers used the collected data to machine learn and create a model that can predict the movements of individual fingers.
Current prosthetic devices for people who have survived amputations or were born with hand aplasia are far from ideal, the study authors note. Many such devices use a primitive control system based on the use of two large muscles in the forearm.
Such prostheses can form one to ten grips, but the movements of individual fingers are not utilized. The new generation of prostheses, based n