The tricky part of connecting a stupid, thinking brain to a cold, working, and inactive computer is getting information through your thick skull ̵1; either mine or anyone else’s. After all, the whole point of the skull is to keep a brain safely separated from it [waves hands at everything].
So if that brain isn’t yours, the only way to know what’s going on inside it is to reason. People make very educated guesses based on what the brain tells the body to do – for example if the body makes some sort of noise you can understand (it’s words) or moves. around in a recognizable way. That’s a problem for people trying to understand how the brain works, and the problem is even bigger for people who are unable to move or speak because of an injury or illness. Sophisticated imaging technologies like functional magnetic resonance can give you some clues. But it would be nice to have something more direct. For decades, technologists have been trying to make the brain communicate with a computer keyboard or robotic arm, to get meat combined with silicon.
On Wednesday, a team of scientists and engineers came up with results from a promising new approach. It involves attaching electrodes to an expandable, expandable tube called a stent and threaded it through a blood vessel leading to the brain. In tests on two people, the researchers actually went into the pipeline, ran a pointed stent wire up that vein in the throat and then into a blood vessel near the main motor cortex of the brain, where they put springs. The electrodes clasp the circuit walls and begin to sense as people’s brains signal their intention to move – and send those signals wirelessly to a computer, via a surgically inserted infrared transmitter into a cage. the subject’s chest. In an article posted above Journal of NeurosurgeryAustralian and American researchers describe how two people paralyzed with unilateral atrophic sclerosis (aka Lou Gehrig’s disease) used such a device to send messages and deceive online only. by controlling the brain.
“Self-dilation stent technology has been well proven in both cardiovascular and neurological applications to treat other diseases. Thomas Oxley, an intervention neuroscientist and CEO of Synchron, which is hoping to commercialize the technology, says we’re just using that feature and placing electrodes on top of stents. “It is completely transplantable. The patient goes home after a few days. And it’s plug-and-play. “
After the subjects went home, they had to practice. Electrode stents can pick up signals from the brain, but machine learning algorithms have to find those signals – the mind’s imperfect reflexes when working even under ideal conditions – really great. for what. But after a few weeks of work, both patients can use an eye-tracking device to move the pointer and then click with thought, using implants. It may not sound like much, but that’s enough for both to send text messages, shop online and carry out the activities of everyday digital life.
The Food and Drug Administration has yet to approve what Oxley calls a “stentrode” for widespread use and the company is still looking for funding for more trials, but these preliminary results suggest it’s a working brain-computer interface. The signals it received did not contain sufficient information. Right now, all the stentrode is collecting is just a little bit of information – a telepathic click or without that click. But for some applications, maybe that’s enough. “There has been a lot of discussion about data and channels, and is it really important that you deliver a life-changing product to your patient?” Oxley said. “With just some restored outputs for patients they’re under control, we’ve helped them take control of Windows 10.”
More ambitious computer-brain interfaces and neural prostheses have surfaced lately. Last month, Elon Musk’s company Neuralink showed off the wireless BCI with more than a thousand flexible electrodes designed to be inserted directly into the brain by a dedicated robotic surgeon. (So far, the company has only shown short-term use in pigs.) The installation of electrodes is difficult; While it is true that brain surgery is not rocket science, it has risks whether the surgeon is a robot or not. Even thin, flexible electrodes like the Neuralink have been shown to be invasive enough for the brain to try to protect against them, coating them with glial cells that reduce the ability to conduct electrical impulses they are looking for. . And while electrodes implanted as the more commonly used “Utah arrays” electrodes can receive clear signals from individual nerve cells, understanding those signals means What is still the scientific issue under study. Plus, the brain froze like a jelly in donuts; The electrodes fixed in place can damage it. But get it right and they can do more than study the brain. “Locked” patients with ALS have used them as a successful brain-computer interface, even though they require training, maintenance, surgery, etc.
Meanwhile, electrodes placed directly on the scalp can receive brain waves – EEG or EEG – but those electrodes lack the spatial detail of the implant. Neuroscientists know, very roughly, what parts of the brain do, but the more you know about which neurons are activating, the more you can know what they are doing.
A more recent innovation, shell power, placed an electrode grid directly on the brain’s surface. Combined with the intelligent spectral processing of the signals received by the electrodes, the ECoG is good enough to convert the action in the motor cortex that controls the lips, jaw, and tongue into text or even speech. . And there are other approaches. The CTRL Lab, which Facebook bought for possibly upwards of $ 1 billion in 2019, tries to get motor signals from neurons in the wrist. The kernel uses its near-infrared spectral function on the head to sense brain activity.
Oxley and his colleagues stentrode, if it continues to produce good results, will lie somewhere along the spectrum between the implanted electrodes and the EEG. Inventors hope to be closer to the first than the second. But those are still early days. “The core technology and the core idea are too great, but based on where they access the signal, my expectation would be that this is a relatively low fidelity signal,” Vikash Gilja said. Other brain-machine interface strategies. who runs the Neurological Engineering Laboratory at UC San Diego. “At least we know that recording high density electrocardiograms from the surface of the brain can transmit information beyond what is presented in this article.”
A possible problem: The tissue conducts electrical impulses, but the electrodes in the stent are collecting signals from the brain through the cells of the blood vessel. That reduces signal content. “If we took those cortical surface records and compared them to the Utah plaque experiments – most of the clinical experience with implantable electrodes – I would say that the pattern in the ECoG is,” Gilja said. a speed limit, ”Gilja said. (Just for the sake of transparency, I should point out that Gilja did pay work with BCI firms including Neuralink, where Synchron could theoretically compete to one day.)
So it may not be good enough for neuroscience, but it can be very helpful for a paralyzed person who wants to have low maintenance BCI without having to drill through the skull. “There is a trade-off between how invasive you want to be and how much you gather information,” says Andrew Pruszynski, a neuroscientist at Western University in Canada. “This is an attempt to go to an intermediate, to put a catheter close to nerve activity. Obviously it’s invasive, but certainly not as invasive as putting electrodes in the brain.
And more work to do. Oxley’s team hopes to expand their research to more human subjects. They will look for possible side effects, such as the possibility that the stent could contribute to the stroke (although this is less likely because it dips into the vessel wall, a process known as is endothelial). They may find better sites for stents, in blood vessels adjacent to other brain regions of interest; Oxley said anywhere within 2 millimeters of a ship large enough to accommodate the stentrode is a fair game. The software could have improved a bit, in terms of understanding the true meaning of the brain when it emits electrical bells and whistles, and some of their tests suggest the system can collect more information. details – like what specific body the user is trying to contract with. That could lead to more useful prostheses or control of devices outside of Windows 10. “The motor system, right now, is what will provide therapy for people with paralysis,” Oxley said. “But when we start interacting with other regions of the brain, you start to see how technology unlocks the brain’s processing power.” It’s hard to predict what might happen when scientists actually figure out how to get inside someone’s head.
This story originally came out wired.com.