Scroll Top

Revolutionary Bio-Compatible transistors will let AI’s interface with your brain



Neural interfaces that let us communicate with each other and the machines around us, as well as stream our thoughts to TV, will revolutionise humanity, but we need bio-compatible hardware.


It’s a fact that silicon transistors and the brain don’t mix, and while you might not think that that’s much of an issue today, since silicon chips are designed to go into computers and smartphones, and not into the human brain, as we race full throttle into the future it is. And it’s an increasingly pressing issue as companies from Elon Musk’s NeuraLink, to Facebook and the US Military, try to find new ways to interface our brains with AI’s, computers, communications systems, and even social networks, and successfully test new ways to upload and download information from our brains – including the ability to read our minds and stream our thoughts to TV’s in colour. Yes, the future is crazy and it’s all being video’d.


Google's AI is now helping chip designers design faster AI chips


Which, all in all, means that fashioning new hardware that conforms to and compliments our biological wetware becomes increasingly important.

To be fair, silicon transistors, when made into electrode arrays, can perform the basics, such as recording neural signals, and processing and analysing them using increasingly sophisticated programs that detect neural patterns, which in turn can be used to stimulate the brain, let ALS or Locked In patients perform Thought-to-Text, and, or control smart Neural-Prosthetics.

The big problem though? They’re not biocompatible in the long term. Without modification, implanted electrodes invariably activate the brain’s immune system, resulting in scar tissue around the implantation site as the cells eagerly attack the foreign silicon invaders.

The trick therefore, at the moment at least, is to encase them in plastics that the human body can tolerate. But if you’ve tried squeezing a sleeve-protective laptop into a small bag, you’ll know that increasing bulk stretches out the bag, and in the case of Brain Machine Interfaces (BMI), brain tissue is the bag, and that can have a negative effect on the patients wellbeing.

To Dr. Dion Khodagholy at Columbia University, the solution isn’t making smaller transistors – we’ve almost hit the limit, now reaching 5nm, 1nm, 0.5nm, and even single atom scale transistors. Rather, it’s to fabricate entirely new transistors that comfortably interface with human tissue, brain or otherwise. This month, the team described a soft, flexible, and Bio-Compatible Transistor that operates on ions, rather than electrons in traditional transistors, in Science Advances.


Researchers build a path to a 200,000 core chip


Because neurons rely on ions for their communication, the new transistors are far more efficient at processing body signals in real time compared to current generation electronics. In a series of tests, the team was able to string multiple transistors together to amplify signals and form logic gates, similar to those used in silicon-based computing.

The devices, which are made of flexible, biocompatible materials, allowed the team to accurately measure EEG “brain wave” signals without requiring additional adhesives, and lowered the contact space between gadget and scalp by five times compared to the usual setup. If that doesn’t seem particularly impressive, the team has only just begun exploring the potential of their ion-drive transistors.

“Our transistor makes communication with neural signals of the body more efficient. We’ll now be able to build safer, smaller, and smarter bioelectronic devices, such as brain machine interfaces, wearable electronics, and responsive therapeutic stimulation devices, that can be implanted in humans over long periods of time,” said Khodagholy.

By making novel transistors, Khodagholy’s team is digging into the very basics of computation – brain or otherwise. In a nutshell, a transistor is a mini electrical component that does two things very well. One, it works as an amplifier to boost input current, which is what hearing aids or microphones rely on. Two, it works as a switch, allowing a small current to trigger a larger one, this is how computer chips work, with their billions of transistors that can store 0s and 1s and each operating individually.


Department of Defense warns the US is falling behind China on 5G


Silicon transistors, even fancy ones modified for biocompatibility, require Ion-to-Electron conversion during their operation. They act as translators to turn the body’s operating language, such as ions that are a type of charged particle, to one that computers use, and most of them are susceptible to water damage and need to be put into a protective casing, which introduces bulk and decreases performance.

Scientists have been able to minimize some of those issues with organic Electro-Chemical transistors, which rely on biocompatible molecules linked to each other to form a “channel” that allows signals to flow through with the help of external electrolytes, which are liquids that conduct electricity. These transistors, however, can’t be individually controlled, making it impossible to build logic gates and circuits, and they’re painfully slow compared to the brain’s ultrafast operations.

To Khodagholy, an ideal transistor for the brain needs four things. One, it’s built from biocompatible and stable materials, two, it’s soft and flexible to avoid mechanical mismatches with the brain, three, it needs high speed and efficient amplification mechanisms that can tease out and boost useful brain chattering from background noise, and finally, it has to have independent gating, in the sense that each transistor can be controlled separately, which allows them to be linked up into integrated circuits.

The team’s answer to bio-transistors is the Ion-Gated Organic electrochemical Transistor (IGT). In a nutshell, IGTs are built from biocompatible material similar to those previously used. However, they have mobile ions directly embedded into the conducting material that makes up the transistor channel. In this way, they no longer rely on external electrolytes, but are themselves the full package for conducting information.


Sony's newest TV has it's own AI brain to improve your viewing pleasure


The secret ingredient? Sugar.

“Sugar molecules attract water molecules and not only help the transistor channel to stay hydrated, but also help the ions travel more easily and quickly within the channel,” Khodagholy explained.

Because the mobile ions are directly in the transistor channels, they don’t have to travel far to modulate the transistor compared to external electrolytes – the typical solution. This makes the IGT respond orders of magnitude faster than electrolyte-gated transistors to changes in external signalling, said study author Dr. George D. Spyropoulos.

Sticking sugar, which provides an ion reservoir, directly into the transistor had another perk – it allowed each transistor to be made independent. Rather than bathing in and sharing external electrolytes, IGTs have the capacity to have their own gates – that is, a membrane that controls whether they’re on or off. In one experiment, the team microfabricated two separate logic gates and confirmed that each operated accurately, performing their intended arithmetic.

This confirms “the scalability of IGT architecture for use as bioelectronics computational modules,” they said.

In another study, the team found that the devices could reliably amplify tiny signals by as much as four-fold. Because neural and other body signals often require multi-stage boosting before they’re accurately picked up and deciphered, IGTs seem perfectly suited for the job.


World first as two quantum computers go head to head


As a proof of concept of IGT’s biocompatibility, the team turned its focus on EEG. Widely used in clinics and labs, EEG picks up brain waves using a cap of electrodes on the surface of the scalp.

It’s not a fun process – the scalp often has to be exfoliated and an adhesive is used to better stick on the metal electrodes, which causes irritation at best and rashes at worst. Hair also gets in the way and muddies signals.

IGT, in contrast, is a dream. Its small size meant that the team could slip it between hair follicles. Its flexibility and bendiness made it possible to slap it straight onto the scalp – no pre-treatments required. In a test that measures brain signals when people are awake with their eyes closed, IGTs reliably and consistently picked up the brain’s activity.

A portable, lightweight EEG device that can be manipulated by hand could also change the future of neurology. But that’s just the beginning. The tiny size of IGTs means it will be possible to apply more devices to smaller areas to measure signals at a finer scale, or stick them into areas normally too small or irregular to accommodate electrodes. Because they’re intrinsically soft, conformable, and biocompatible, they can be used on extremely delicate tissue, such as a newborn’s scalp or inside the brain.


This ingestible smart pill monitors your gut health from the inside out


But most importantly, said study author Dr. Jennifer Gelinas, IGTs can perform circuit computations. This means they could one day be part of a closed-loop system capable of detecting the brain’s electrical patterns and stimulate accordingly, with far less risk than current electrode-based interfaces which could mean getting rid of those big and bulky blue skull caps you see people wearing so often in “brain experiments.”

“With such speed and amplification, combined with their ease of microfabrication, these transistors could be applied to many different types of devices. There is great potential for the use of these devices to benefit patient care in the future,” said Khodagholy.

Related Posts

Leave a comment


1000's of articles about the exponential future, 1000's of pages of insights, 1000's of videos, and 100's of exponential technologies: Get The Email from 311, your no-nonsense briefing on all the biggest stories in exponential technology and science.

You have Successfully Subscribed!

Pin It on Pinterest

Share This