2023-07-10 17:30:00
“People who have suffered neurotrauma, such as stroke or brachial plexus injury, often see reduced limb functionality,” said researchers from Florida Atlantic University (USA). Clearly, daily tasks can be extremely difficult due to decreased coordination and strength in one or both upper limbs. Faced with these problems, scientists have developed connected devices to improve the capabilities of patients. This is the case of the Florida team.
Bionic gloves made of soft, flexible materials and sensors
As part of a study, she developed a pair of bionic gloves to help pianists who have suffered a stroke relearn how to play this instrument by “feeling” the difference between correct and incorrect versions of the same song. These connected gloves, which adapt to the user’s hand, contain special sensor networks connected to each fingertip of the robotic glove.
According to the authors, the artificial intelligence provides precise force and guidance to recover the fine finger movements needed to play the piano. By monitoring and reacting to users’ movements, the bionic glove, designed from 3D-printed polyvinyl acid stents and a hydrogel cast, provides real-time feedback and adjustments, which makes it easier for them to acquire the correct movement techniques.
“Playing the piano requires complex, highly skilled movements, and relearning tasks involve restoring and retraining specific movements or skills. Our Bionic Glove is made of soft, flexible materials and sensors that provide support and assistance gently to people who relearn and regain their motor skills”, said Erik Engeberg, lead author of the work and professor at Florida Atlantic University, in a statement.
More than 94% accuracy thanks to the ANN algorithm
To test the effectiveness of these connected gloves, scientists programmed them to feel the difference between correct and incorrect versions of the tune from “Mary Had a Little Lamb” played on the piano. To introduce variation in performance, they created a set of 12 different types of errors that might occur at the start or end of a note, or due to premature or delayed timing errors, and which persisted. for 0.1, 0.2 or 0.3 seconds.
The results, published in the journal Frontiers in Robotics and AI, showed that the Artificial Neural Network (ANN) algorithm, used to classify song variations, had the highest accuracy, at 97.13% with a human and 94.60% without a human. He managed to determine the error percentage of a certain song and identify the keys pressed outside the time limit. “These data underscore the potential of the smart glove to help people with disabilities improve their hand dexterity,” concluded the team.
1689011642
#connected #gloves #relearn #play #piano
.jpeg?fit=%2C&ssl=1)
