Brain-Computer Interfaces and Well-Being

The brain-computer interface (BCI) sector is thriving. As with many new and high-profile technologies, fully visualizing the potential applications for these developments may be difficult. To gain a deeper understanding of the practical uses for BCIs, Mouser Electronics spoke with Matt Angle, Chief Executive Officer and founder of Paradromics.

Understanding the Technology

We started our conversation by asking Angle about any limiting factors to how BCIs function now. He explained that "BCIs are currently dictated by the amount of data we can push across the interfaces. The state of the art in BCI research today is technology that was developed in the late 1980s. As a result, the effective decoding rate with this technology tops out around 10 bits per second (BPS). These devices can do useful things in the clinic, but they could do much more. Imagine trying to stream an HD Netflix movie over 1990s dial-up internet—that is the state of BCI hardware today."

Angle continued: "But now we have well-financed companies that approach the problem like tech companies rather than research device companies, and so we are going to see a pipeline of new devices and new applications coming out of the world of BCIs. We're going to see BCIs starting to develop in the same way that the early internet did."

Angle views Paradromics as unique in the BCI world, "We're building high-data-rate BCIs that are going to be reliable for more than 10 years. The grand design challenge of building next-generation BCIs is combining performance with reliability. [Many] labs across the world can build high data rate systems that might last for a few months. [Many] medical device companies create devices designed to be stable for 10 or 20 years. At Paradromics, we make an implantable electronic device that will function safely for more than 10 years, talking to individual neurons, capturing data, and transmitting them out of the body."

Targeting Populations of Neurons

As Paradromics captures these data from the brain's neurons, the information gained is additive. That is, the more neurons technicians record from, the greater their understanding will be. However, this method requires proximity to the neurons themselves, which is difficult to achieve from the surface of the brain.

As Angle explained, "The best analogy we have is a stadium full of 100,000 people. If you drop microphones into the bleachers, every microphone will record a conversation. If a few people are around each microphone, you will be able to distinguish their voices and capture a large amount of information. Every microphone you drop in is more information. We detect data in the brain in the same way, by placing electrodes close to the neurons."

"The alternative is to record the surface activity of the brain," he continued. "In our analogy, this is like placing a microphone in the parking lot outside the stadium. Rather than capturing individual conversations, all you will hear is the roar of the crowd. Perhaps machine learning tools will be able to tell which team has scored based on changes in the noise of the two groups of fans, but you will never be able to [discover] what spectators think of each play."

To get closer to the brain's neurons, Paradromics employes a specialized BCI device. "Paradromics' direct data interface (DDI) uses an array of electrodes, each 40µm in diameter, that reaches down about 1.5mm into the cortex. If you look closely, it resembles a tiny brush with electrodes that are almost too small to see."

Such a delicate device could present problems when installed, but Angle was clear on the advantages of such fine electrodes: "The size of the electrodes is so important, as these little hairs do not disrupt the blood vessels very much. A comparison would be putting a finger through a woven blanket. The fibers of the blanket part to allow the finger to pass through. However, if the electrodes were larger—more than 50µm—the insertion process becomes far more disruptive, akin to trying to push a water bottle through the same blanket. The disruption to the blood vessels could trigger an immune response, affecting the long-term reliability of our product."

While the placement of the electrodes is deliberate, it is not required to be precise. "We do not have to target specific neurons," Angle explained. "The electrodes are inserted down into a population of neurons, which work together to encode information, so we simply need to sample the population randomly."

Introducing New Applications

According to Angle, the best practical application of Paradromics' DDI technology is speech. "Arguably, this is the most complex task that is being accomplished with BCIs. From previous research, we know which areas of the brain control movement. With electrodes in the right places, we can record someone's intention to move the hand. Replicating speech uses the same technique—it is just a motor decoding task. We target the parts of the brain that control the lips, the tongue, and the larynx so we can identify what the patient is trying to do with the body, and then we map that into speech."

"The next step is the training process," Angle continued. "The patient may be paralyzed and unable to speak but is asked to attempt to say words prompted by a computer. After several attempts to speak, the machine learning algorithm starts to be able to predict what the person is trying to say just based on the brain activity. The training process may take as little as a few hours. The result is someone [previously unable to speak] being able to communicate easily with loved ones."

When asked whether he saw his products as medical technology or if they will enter the consumer market, Angle provided a refreshing perspective on the use of BCIs: "We need to think of BCIs in the same way that we view innovations that really made quality of life better for humans—things like refrigeration, antibiotics, fresh water, and vaccines. Just providing a device that connects us with AI or other technology is not the big story." 

Angle continued: "Instead, BCI devices have the potential to transform the quality of life for people suffering [from] a range of conditions. If we can understand how the brain works for someone with treatment-resistant depression or bipolar disorder, we will be able to create effective tools that allow that person to manage the condition. That is far more valuable than the ability to control one's smartphone with one's brain."

Finally, Angle shared with us a very optimistic view about the future of BCI technology: "I think that BCI [technology] is on the cusp of doing something for mental health like the continuous glucose monitor did for diabetes. I would encourage people to be even more ambitious about what BCI [technology] could do, because I think that it could be a way of tackling some of the biggest challenges facing us today in terms of mental health, well-being, and contentment."