One of the main problems with robotic hand implants is sending the command from the patient’s brain to the hand and fingers, through the muscles or nerves, to make the necessary movements.
It is an obstacle that seems to be solved by a new project carried out by researchers at the University of Michigan who have developed a system to control with extreme precision a robotic hand that responds to the patient’s thoughts.
The interface is based on a series of small muscle grafts that amplify the nerve signals that leave the brain. The interface has already been tested on robotic prostheses and seems to have been successful after translating the signals from the brain into real movements with a stable level of precision over time.
Current robotic prostheses work mostly by recording electrical signals from the muscles. This means that the patient has to contract the muscles in the forearm to control the fingers. This is not possible for all patients: in some of them, the muscles may be damaged or completely missing and patients cannot do the necessary push-ups to start the input.
In addition, if the person’s arm sweats, swells or the prosthesis moves, the signal is no longer accurate and the device needs to be recalibrated so much so that in the end many amputees decide that it is not worth it.
The new devices are based on nerves. They also transmit the commands of the brain along the arm. However, implanting wires into the nerves to capture their electrical signals is not efficient because the nerve signals are weak.
It has therefore been thought in the past to increase these signals from the nerves by connecting them to the muscles. In practice, the weak nerve signals have been redirected to another muscle in the body. But even this approach has in many ways proved to be unsuccessful.
Researchers at the University of Michigan, led by Paul Cederna, have therefore thought of an alternative approach, namely to equip the nerves with their own small muscles.
They connected the nerves of the arm to pieces of muscle tissue collected from the thigh, creating a new set of finger muscles inside the forearm. In the study, published in Science Translational Medicine, researchers describe the first tests with this new robotic prosthesis in which wires are inserted through the skin into alternative muscle grafts.
During testing, the device seemed to be able to easily pick up electrical signals. Using computer algorithms, the researchers were able to translate electrical signals into expected movements during tests in which patients moved a virtual hand on the screen or through a commercially available prosthesis called LUKE.
Now researchers are trying to compact this interface by remodeling it so that no wires protrude through the skin and optimizing its functions.