Two “groundbreaking” brain-computer interfaces are being tested on paralysed people who lost the ability to speak.
Category
🗞
NewsTranscript
00:00 [Music]
00:23 Signals from the individual brain cells or neurons are amplified, sent out through a pedestal,
00:30 which is a stainless steel device that's placed on the patient's skull.
00:35 It comes out through the scalp.
00:37 The signals are then amplified, run through a machine learning software algorithm,
00:45 which associates the neural activity, the brain activity,
00:49 with the individual speech sounds that are flown into English.
00:53 [Music]
01:04 The reason that we like these is that, yeah, they don't penetrate into the brain,
01:07 and they cover a larger region of the brain, so it's possible to record from more areas,
01:13 and the recordings are more stable, meaning they look the same, you know, day to day,
01:20 since they sit on top of the brain.
01:22 And they're also recording from tens of thousands of neurons at once,
01:27 whereas electrodes will record from, you know, one to three neurons at once.
01:31 So it's a different signal that still has a lot of information.
01:35 [Music]