We imagine that putting on a prosthetic limb for the first time ever can be rather disconcerting. This means that patients tend to take time to get used to it, to walking around, and potentially even running. However it seems that researchers believe that by using AI to tune these prosthetic limbs, it could help speed up the recovery process.
In a paper published in the IEEE Transactions on Cybernetics journal (via VentureBeat), researchers at the North Carolina State University and the University of North Carolina talked about a system that can apply “reinforcement learning” when tuning a prosthetic limb, such as a robotic knee.
What happens is that the AI will take into account various aspects of the limb, such as how stiff the joint is, or the range of vertical motion allowed in a foreleg, and as such will tune the limb accordingly which should allow the wearer to have a more comfortable experience. In some tests, the system managed to help a patient walk naturally on level ground in just 10 minutes.
However the drawback to the system in its current state is that it does not know if the changes and adjustments it makes are for the better. According to the paper’s co-author and professor of biomedical engineering at both North Carolina State University and the University of North Carolina Helen Huang, “If you wanted to make this clinically relevant, there are many, many steps that we have to go through before this can happen,” she said. “So far it’s really just to show it’s possible — by itself that’s very, very exciting.”