In a survey conducted in 2020 among upper limb amputees, 20.7% used myoelectric prostheses and 74.4% used devices with hooks that open and close through body movement and a pulley system. No differences were observed in the degree of user satisfaction between the different types of prostheses and, of those who reported having used myoelectric prostheses, 44% abandoned them completely.
Phantom aims to interact directly with muscles to give users “more natural” control, but the company had to test a wearable model of its muscle sensor before it could implant the device in volunteers. Although the portable version worked well, Glass claims that it is not ideal for daily use. Like existing prostheses, it can slip and move, affecting its accuracy. That’s why you have to recalibrate it often. According to the CEO, an implant would be more reliable and could achieve even greater precision, since it would not have to read electrical signals through the skin. The implantable version is expected to be inserted through a small incision and inserted under the skin.
“We collect electrical activity directly from the surface of the muscle. In amputees, those neural pathways are still intact,” he explains. The intention to move originates in the brain, which sends electrical impulses through peripheral nerves to signal the muscle to contract.
Preparation before the clinical trial
For the wearable study, Smith and the other study participants had a one-hour training session to familiarize themselves with the technology and returned a second day to take the test. After a 10-minute algorithm calibration process on the day of testing, participants were instructed to make 11 gestures: such as opening the hand, making a fist, pinching, raising the thumb, pointing with the index finger, clicking with the index finger, rotating the wrist inward and rotating the wrist outward several times each while the software Phantom learned and decoded his muscle signals in those real movements. For able-bodied participants, this meant that a robotic arm imitated their gestures. For Smith, the prosthesis made the movements he simply intended to make: “It was a cool experience.”
Participants wore two thin sensors, each with 16 electrodes. Gesture decoding accuracy ranged between 84.8 and 98.4% across participants, with latency; the time that elapses from the detection of the signal to the execution of the gesture, less than 200 milliseconds. Natural human latency is about 100 milliseconds. “The speed of your ability to achieve those classification accuracies is enormous. The faster you can do that, the better off you are and the more fluid the system can be,” says Paul Marasco, a neuroscientist at the Cleveland Clinic who studies touch and natural movement for artificial limbs, but does not participate in Phantom.
Science Corporation’s retinal implant allowed some people who had lost central vision to read, play cards, and recognize faces.
The market for brain-computer interfaces
Some companies are developing brain implants that will allow paralyzed people to control their prostheses with thought. These systems, known as brain-computer interfaces (BCI), read and decode signals from the brain so that the prosthesis performs the desired movement. Neuralink, the startup Elon Musk’s CEO announced last month that he will launch a study to test whether his brain implant can allow participants to directly control a robotic arm.
Geoffrey Ling, Phantom’s technical advisor and founding director of the Biological Technologies Office at the US Defense Advanced Research Projects Agency, suggests that brain implants carry more risks and must be very durable so that patients do not have to undergoing multiple brain operations: “Peripheral nerves are a very attractive approach because it is minimally invasive.” Phantom believes its implant could be inserted during an outpatient procedure without the need for a specialized surgeon.
By 2025, Phantom plans to begin a clinical trial of its implanted version, which will involve upper limb amputees.; Smith hopes to participate in that study. If the company’s technology becomes commercialized, it could help amputees like him perform everyday tasks more easily. “I think this is going to totally change the rules of the game,” he rattles off eagerly.
Article originally published in WIRED. Adapted by Alondra Flores.
#Mindcontrolled #muscle #implants #alternative #embedded #brain
