Article The brain-implant company going for Neuralink’s jugular (design, engineering)

[C C]

https://spectrum.ieee.org/synchron-bci

INTRO: For more than half a year, six people have been going about their lives with sensors implanted in blood vessels in their brains that enable them to communicate directly with their computers. The participants, who are all severely paralyzed, are taking part in a study that could change their lives and mark a turning point in brain-computer interface (BCI) technology. In 2024, they’ll find out if the tech will continue on the path to the clinic.

Until now, only about 50 humans have ever had BCIs implanted in their brains. And only a handful of those people have been able to leave the laboratory to use them in the real world, since most BCI implants involve wires protruding from the head. The new study is the largest human trial of a fully implantable, at-home BCI system.

And no, the maker of this device isn’t Elon Musk’s Neuralink. It’s a company called Synchron, and it is quietly leading the race to bring a BCI implant to market.

Synchron’s Stentrode is inserted through the jugular vein and snaked up to a blood vessel over the motor cortex. Synchron

“Synchron is the very first to commercialize the concept of BCI [implants] in a meaningful way, and they’re paving the way for the whole field,” says Nick Ramsey, a clinical neuroscientist at University Medical Center Utrecht, in the Netherlands, who is not involved in the development of Synchron’s device. It “might very well be on the market for a while before any [other] devices are competing with it,” he says.

If Synchron’s system works, it will provide an invaluable communication method to people with severe paralysis... (MORE - details)

[Yazata]

Oww! That touched a nerve!

They discuss three companies working in this field, each seemingly with its own strength and weaknesses.

One big advantage of the Synchron device: It doesn’t require open brain surgery. Instead, it’s delivered like a stent. A 16-electrode array, trailing a lead behind it, is inserted into the jugular vein in the neck and snaked up a blood vessel near the brain’s motor cortex. When it reaches its destination, it springs out into a tubelike scaffold that fits against the inside wall of the blood vessel. There, the Stentrode records electrical activity coming from nearby brain tissue. The Stentrode is connected by the lead to a small receiver-transmitter in the chest, which wirelessly sends data to an external digital device.

It looks like the plusses here are simplicity and minimal invasiveness. The negatives are how localized the Stentrode seems to be, with all of its electrodes located in a single blood vessel.

The only computer commands Synchron’s system can currently generate are clicks and a scrolling function. The click command can be used in conjunction with special assistive software that slowly scrolls through Web pages highlighting different areas of interest that the user can click on. Synchron is limited to these two commands because of the quality of the brain signals that can be detected from inside a blood vessel. “We are recording population-level signals from neurons, not single-unit-level signals,” says Tom Oxley, CEO of Synchron.

Certainly better than nothing. But compare that to Neuralink's monkeys which effortlessly played video games just by thinking. Actually faster than a player with a hand controller.

This setup is considerably less sophisticated than the BCIs coming from other companies and academic groups. “The current research is geared toward capturing more signals from a larger patch of cortex so that you are decoding more complex movements or speech,” says Ramsey, who is credited with testing the very first implantable BCI in a human in 2016. “To decode speech, you need at least 100 electrodes.”

Synchron’s competitors include Elon Musk’s Neuralink and rival Precision Neuroscience. Neuralink’s BCI features 1,024 electrodes distributed across 64 ultrathin threads, which must be surgically implanted by a custom robot. The company has tested it in animals and in May 2023 said that it had received a green light from the FDA to test it in humans, after the agency initially rejected the request.

Precision Neuroscience’s BCI features 1,024 electrodes on a one-square-centimeter flexible film. In 2023 the company conducted pilot studies in humans in which the film was placed temporarily on the surface of the brain while people were undergoing unrelated tumor surgery.

Neuralink's electrode implant is three dimensional and is implanted deep into the motor (and eventually the sensory) cortex in the form of tiny wires much thinner than a human hair. Precision Neuroscience has the same number of electrodes, but they seem to be a two dimensional set that lies atop the surface of the brain instead of inside it. So it remains to be seen whether it can achieve resolutions comparable to what Neuralink can achieve.

The 2-D approach will be better (easier to install and less invasive) provided it's equally effective. But my guess is that it won't be.

So...

Synchron - advantages are simplicity, least invasive (requires no surgery) and quickest into the clinic.

Disadvantage is the it will probably prove least effective.

Neuralink - advantage is that it at least has the potential of being most effective at restoring lost function and eventually sensation as well. Elon has discussed providing Teslabot style arms and legs to amputation victims complete with joint position encoders and touch sensors. Perhaps synthetic vision for the blind. Or perhaps even (eventually) remote telepresence where one perceives one's self inhabiting a robot body. Quadriplegics or locked-in individuals could animate an Optimus-like robot body while its sensors feed in through the person's sensory cortex, getting them back out into the world. (Great for soldiers on the battle field. If a bad-ass military robot is blown up the remote operator needn't feel pain and emerges uninjured.) More speculatively, it's possible to imagine whole virtual universes in which implanted individuals perceive themselves as virtual avatars in a fully-immersive computer- generated video-game environment as if it was real, Ready Player One style.

Disadvantage is that it's far more complex and surgically invasive. The skull has to be opened and many electrodes precisely implanted. It will take longer to receive regulatory approval and find its way into the clinic. And some of the more sci-fi applications I outlined above might depend on advances in neuroscience that remain to be made.

The Precision Neuroscience approach seems to occupy the middle ground in those two approaches. It's simpler, less invasive and probably faster into the clinic than Neurolink, but more complex, more invasive (it requires opening the skull) and probably slower into the clinic than Synchron.

Advantage/disadvantage is that it will probably be less effective than Neurolink's approach. But it might prove to be better than Synchron's fast and (relatively) easy approach.

While it's easy to think of these three companies as competitors, I think that they might better be conceived as a synergetic three-fold approach to the very real problems that many people face. They can all do good and they probably all have their niche. I expect Synchron to hit the clinic first and do a lot of good for a lot of people. Then Precision Neuroscience and eventually Neurolink will become the leaders as we see the basic approach growing, maturing and improving in ways that are hard to imagine today.

[Yazata]

https://www.youtube-nocookie.com/embed/CxBbw8eT624