Advertisement

Software tunes powered prosthetic legs automatically

The system can adjust based on what a person is doing, be it walking or carrying luggage through an airport.

By Stephen Feller
Researchers developed software that allows powered prosthetics to tune themselves automatically, making the devices more functionally useful and lowering the costs associated with powered prosthetic use. Photo by Helen Huang/NC State and UNC-Chapel HIll
Researchers developed software that allows powered prosthetics to tune themselves automatically, making the devices more functionally useful and lowering the costs associated with powered prosthetic use. Photo by Helen Huang/NC State and UNC-Chapel HIll

RALEIGH, N.C., Sept. 29 (UPI) -- Powered prosthetic legs are programmed to mimic a person's natural movement while walking, but changes in weight or the way a person moves can require them to be reprogrammed for ideal function -- which can be inconvenient and costly.

Researchers have written an algorithm that can be used with the software in powered prosthetics to automatically adjust the leg based on what a person is doing, be it walking up stairs or pulling luggage through an airport.

Advertisement

"When a patient gets a powered prosthetic, it needs to be customized to account for each individual patient's physical condition, because people are different in size and strength. And that tuning is done by a prosthetist," said Helen Huang, an associate professor in the biomedical engineering program at North Carolina State and the University of North Carolina Chapel Hill, in a press release. "In addition, people are dynamic -- a patient's physical condition may change as he or she becomes accustomed to a prosthetic leg, for example, or they may gain weight. These changes mean the prosthetic needs to be re-tuned, and working with a prosthetist takes time and money."

Advertisement

Researchers developed a cyber expert system, or CES, that encoded tuning decisions normally made by prosthetists as computer rules in the auto-tune algorithm. The tuning performance of the software was tested on four men, two able-bodied and two trans-femoral amputees.

In the tests, the prosthetics improved as the men took more steps and the software adjusted to the way they walked, although not as well as natural limbs.

The system obviously does not include the experiential knowledge of a professional who can consider the way an adjustment will effect a person's back while walking, however the algorithm can more precisely set a proper joint angle -- so it is a work in progress.

"So we know our approach works," Huang said. "Prosthetists rely on years of experience to not only adjust the joint angle, but to adjust a prosthesis to help patients maintain a comfortable posture while walking. We're not yet able to replicate the prosthetist's success in achieving those comfortable 'trunk motions,' but it's something we're working on."

The study is published in Annals of Biomedical Engineering.

Latest Headlines