Most of us go through the day without stopping to think about the remarkable functionality our limbs provide, and the many tasks they allow us to do. When someone loses a limb due to injury or illness, a prosthetic limb can restore some of the capabilities lost, but the dexterity of a biological limb usually cannot be matched. For jazz musician Jason Barnes, who lost part of his right arm in a work accident, this meant losing the ability to play the piano.

Now researchers at the Georgia Institute of Technology have created an artificial intelligence (AI) prosthesis that helps restore Barnes’ ability to play. Like most typical prosthetic limbs, Barnes’ everyday prosthesis is controlled by electromyogram (EMG) sensors attached to the muscles, which detect electrical impulses. Flexing his forearm muscle allows the index finger and thumb to clamp together; contracting the muscle closes his fist. While EMG sensors recognize muscle movement, they are not capable of determining which individual finger wants to move because the signals are too noisy. The AI prothesis, however, gives Barnes the ability to control each finger individually. Georgia Tech professor Gil Weinberg, who lead the research, used an ultrasound device to see muscle contractions and the speed and direction of muscle movements. By attaching an ultrasound probe to the arm, Weinberg was able to train a deep learning network to analyze and detect muscle movements. The research team then created an algorithm that can predict which finger the musician is trying to use. As Weinberg pointed out, “By using this new technology, the arm can detect which fingers an amputee wants to move, even if they don’t have fingers.”

This improvement in prosthetic technology has the potential to enhance an amputee’s life in other ways; it could be used for fine motor activities such as bathing, holding a fork, and grooming, which are challenging with standard prosthetics. While biological limb function might be the gold standard, a prosthetic powered by an ultrasonic sensor that allows fine motor hand gestures is a powerful advance in prosthetic technology.