Machine Learning is Improving The Way These Prosthetic Limbs Perform - Geek.com

Prosthetic limb users may have a bright future to look forward to, in terms of new technology that can and will help them adapt to the challenges of using the tools. AS it turns out, according to Tech Xplore, the scientists at Imperial College London and the University of Gottingen have successfully utilized machine learning to help improve the performance of prosthetic hands. The team tested their prototypes on five amputees, and found that the new form of machine learning-based control was actually more adept at providing natural movements than what's currently available to users at this time. This research and its related findings, could potentially help to spark a "new generation of prosthetic limbs," according to the researchers. Professor Dario Farina, who authored the research paper from Imperial College London's Department of Bioengineering, stated that the main goal when designing "bionic" prosthetic limbs is to let patients control them as naturally as possible, as if they were actually their natural limbs, and this new machine-learning technology helps achieve that state as closely as possible, right now. The new bionic hand that's being worked on relies on a human-machine interface that actually helps interpret the user's intentions and sends various commands to the artificial limb, as you can see demonstrated in the video below.