Three Year Study Hopes To Improve Upper Limb Prostheses

Researchers in the Department of Industrial and Systems Engineering at Texas A&M University are studying machine learning algorithms and computational models to gain insight into the mental demand placed on upper limb prosthesis users, with the ultimate goal of improving the current interface in prostheses.

Led by Maryam Zahabi, PhD, an assistant professor, the team is studying prostheses that use an electromyography (EMG) based human-machine interface. The electrical activity generates signals that trigger the interface, translating the signals into unique patterns of commands to allow the user to move the prosthesis. The Texas A&M team is collaborating with researchers from North Carolina State University and the University of Florida.

The three-year project, funded by the National Science Foundation, will focus on a combination of cognitive performance modeling, machine learning algorithms, and virtual reality and driving simulations to identify the cognitive workload of EMG-based prosthetic devices. “Our models can provide information regarding why certain technologies may lead us to higher or lower workload, which will ultimately help in providing engineering and design guidance for less mentally demanding prosthetic devices,” said Zahabi. “This project investigates cognitive workload of prosthetic devices and how it impacts psycho-motor performance and patients’ interactions with these technologies.”

The researchers have hypothesized that the greater cognitive demand in prosthetic device use will lead to reduced motor learning potential and reduced skill retention based on device training. While they are in the initial phases of the study, the results from the researchers’ prior studies in this domain showed that a prosthetic device with pattern recognition based controller was more intuitive, reduced cognitive load, and was more efficient to use as compared to the direct control mode.

“The outcomes of this research can increase accessibility and utility of prosthetic devices for supporting fundamental motor skill rehabilitation or performing activities of daily living,” said Zahabi. “Using the findings of the current project, we hope to investigate the possibility of customized interfaces for prosthetic users accounting for specific user needs in our future studies.”

Source: The O&P Almanac