Biomedical engineers at Newcastle University (
www.ncl.ac.uk/press/news) have created a bionic hand that can “automatically” pick up objects without conscious thought by the user, based on images from a camera.
The researchers claim that their technology acts “just like a real hand”, without the time-consuming processes used by traditional prosthetics.
Current models require users to see an object and physically stimulate muscles in their arms; electrical signals are then picked up and control the hand’s motion, in a process that takes practice, concentration and time.
The new bionic hand uses the built-in webcam and computer vision to automatically analyse objects in front of the user and work out the appropriate grasping action (from a set of four).
The prosthetic then starts picking up the object within milliseconds.
The team, led by Kianoush Nazarpour and Ghazal Ghazaei, showed the computer controlling the hand webcam pictures of many objects at different angles. They then used neural networks so that the ‘machine’ could ‘learn’ the best way to pick up these objects — and similar ones.
Dr Kianoush Nazarpour, senior lecturer in biomedical engineering at Newcastle University, said: “Using computer vision, we have developed a bionic hand that can respond automatically — just like a real hand.
“The user can reach out and pick up a cup or a biscuit with nothing more than a quick glance in the right direction.
“Responsiveness has been one of the main barriers to artificial limbs. Now, for the first time in a century, we have developed an ‘intuitive’ hand that can react without thinking.”