If a sensor was implanted in your brain to connect your brain to the outside world, would you do it? There is such an opportunity now. A few days ago, Musk’s brain-computer interface company Neuralink announced that it has begun recruiting patients for human trials. The main purpose of this PRIME research project is to evaluate the performance of Neuralink's series of brain-computer interface devices on the human body.
The registration method is also very simple. Just go to Neuralink’s official website and click on patient registration.
However, not everyone can participate in this experiment.
Neuralink's announcement makes it clear that people with quadriplegia due to cervical spinal cord injury or amyotrophic lateral sclerosis (ALS) may be eligible for the trial.
No matter how bad it is, you still need to have disabilities in vision, hearing, etc.
The most important thing is that you must be a U.S. citizen over 18 years old.
Under Neuralink's tweets, these volunteers who volunteered suffered from some mild or severe physical diseases and hoped to experience the world again through Neuralink's brain-computer interface products.
If the human experiment is really successful, then as Neuralink said, humans can use their thoughts to control external devices. Not to mention the disabled, maybe we ordinary people are not far away from turning on lights and making coffee with our thoughts in science fiction movies.
However, let’s not dream too soon. The brain-computer interface is not far away from us, but it is not necessarily close to us.
Neuralink's human trial lasted for six years. It is not clear whether it can be used as a horoscope, and the development level of the entire brain-computer interface industry is probably far less mature than we imagined.
When mentioning the brain-computer interface, the image of a chip implanted in the back of the head and connected to a bunch of wires may immediately come to everyone's mind.
Let me give you a brief introduction here. Brain-computer interface can be divided into three types: invasive, semi-invasive and non-invasive.
What Neuralink does is an invasive brain-computer interface.It sounds quite scary to punch a hole in your brain and put this chip that looks like a coin into your brain.
But in fact, invasive brain-computer interfaces in a broad sense have been used in the medical field for a long time. For example, DBS (deep brain stimulation) implants electrodes through minimally invasive nerve surgery to give your nerves a boost from time to time. It is quite effective in treating epilepsy and Parkinson's disease.
But the biggest difference between DBS and Neuralink is that the latter requires a craniotomy.
The risk factor rises quickly!
Of course, if you don’t want to open the skull, there are non-invasive and semi-invasive methods.
Data show that non-invasive brain-computer interfaces account for 86% of the brain-computer interface market size. Now most domestic scientific research institutions and commercial companies basically follow this route.
EEG caps and smart prostheses can also be classified as non-invasive products, which are relatively common in medical rehabilitation scenarios.
But semi-invasive research is relatively rare, and Neuralink’s rival Synchron is one of them.
What they make is this semi-invasive vascular stent, which does not require a cranial opening. It is implanted from the jugular vein, along the blood vessels, to the cerebral cortex to collect signals, and then transmits the data to the outside of the body through an antenna buried under the chest.
The advantage of this solution is that it is less risky than stabbing a hole in the head.
Therefore, in 2021, Synchron obtained clinical approval from the FDA (U.S. Food and Drug Administration) ahead of Neuralink.
Hey, here comes the problem again.
Since both methods can collect neural signals, why do you have to go to the trouble of drilling a hole in your brain? Isn't this just looking for abuse?
Let’s make it clear first that the key to brain-computer interface research is to analyze the collected signals to see what is going on in your mind.
For example, you want to eat hot pot now. (This hotpot is not that hotpot)
Your brain will first form a neural signal of "I want to eat hot pot." The electrodes will capture your signal and then analyze it. Oh, it turns out you want to eat hot pot.
However, the non-invasive method passes through the skull and the semi-invasive method does not enter the cerebral cortex. The interference of noise will cause the collected neural signals to be less clear.
Maybe you wanted to eat hot pot, but it interpreted it as wanting to eat snail noodles, or it simply couldn't interpret it.
Therefore, invasive brain interfaces like Neuralink have always been the "Mount Everest" that is difficult to climb in the industry.
After so many years, sleep monitors, sleep aids, attention training headsets... even if high-quality neural signals are not collected, some people rely on these to sell non-invasive brain-computer interface products, live in villas, and drive luxury cars.
But Intrusive seems to have never heard any big noise.
It can only be said that after the differentiation of technical paths, brain-computer interfaces will die from drought and drought.
So why is invasive brain-computer interface so difficult?
Last year, the FDA conducted a soul torture on Neuralink: Is your brain-computer interface device safe? What should I do if the lithium battery leaks electricity in my brain? After putting the electrode in, how do you take it out? What should I do if the wires move in my brain? ...
I was so frustrated with Neuralink that I had nothing to say.
After all, Neuralink was indeed suspected of animal cruelty due to testing. According to Reuters, Neuralink killed about 1,500 animals including sheep, pigs and monkeys since the beginning of the experiment.
The FDA’s consideration is a practical issue that Neuralink and the entire invasive brain-computer interface have to face.
First is safety,
To implant and remove electrodes, you have to open a cranium, right?
Since the craniotomy scene is too bloody, it is not shown here. Curious friends can search for it on their own.
Its risk factor is not at the same level as double eyelid piercing or laser injection on the eyes.
So Musk moved out a surgical robot called "R1" last year, which provides one-stop services including locating the implant location, removing the skull, implanting the chip and suturing the wound.
The whole process may only take 15 minutes.
The bad reviewer speculates that R1's contribution may be indispensable for Neuralink's previous approval.
Secondly, after the electrode is implanted, it must be ensured that it does not move or leak electricity.
If robots can be used in craniotomy surgery to improve the success rate of surgery, then many things after electrode implantation may focus on a Buddhist approach.
BrainGate, a company in the United States that also makes invasive brain-computer interfaces, encountered a situation where electrodes were scrapped in the brain.
It's not because there's no electricity, it's because the electrodes have entangled the glial cells...
To make matters worse, if a traditional "Utah Array" electrode is implanted, a needle tip that is too hard may cause intracranial infection or rejection.
Immune system: What level do you belong to, staying in the same body as mine?
Furthermore, high-quality neural signals are not always available as much as you want.
The "Utah Array" mentioned above can only transmit neuron signals from 96 electrode channels.
What is this concept?
According to "Moore's Law" in the brain-computer interface world, it will take until 2100 to record one million neurons at the same time, but an adult's brain has about 86 billion neurons...
If you want to collect as many neural signals as possible, you can only put several of these needle-like things in your brain, and the risk comes up again.
Therefore, in the past two years, many scientific research institutions have been tinkering with flexible electrodes with more electrode channels. It is said that they can change shapes with nerve cells, but we have not seen any big waves yet.
For example, Neuralink developed a flexible electrode that can "seamlessly fit" with nerve cells in 2019, and also expanded the number of electrode channels to 1,024.
Although it cannot perfectly solve the problem of neuron signal transmission, it at least looks much safer than the previous "Utah Array".
What's more, Neuralink passed FDA clearance in May, which means that invasive brain-computer interfaces are feasible in terms of safe operation.
Maybe the next time we see news about Neuralink, it will be that a certain patient with ALS or depression has recovered from health with the help of the brain interface.
Maybe in the future, the bad reviewer will directly code the words and publish the article (dog head).
Finally, let’s think about it again. If the technology is fully mature one day in the future, what will you use the brain-computer interface for?