Featured Founder: Paradromics Founder and CEO Matt Angle
What inspired you to start Paradromics?
As a researcher at Stanford, my colleagues and I were working on a way to scale up electrical recording in the brain to better understand how large networks of neurons encode information. We were looking at architectures for neural recording that could scale to 10,000+ neurons simultaneously.
At the same time, clinical trials across the country were showing that electrical recordings in the brains of people with paralysis could allow them to control computers and robotic arms. Those brain computer interface [BCI] demonstrations were incredible, but their performance was data-limited, with only 100 electrodes recording from about the same number of neurons! Plus, these systems were confined to the lab and required the patient–and their brain implant–to be directly tethered to a computer with cables.
We knew that if we brought our neural recording technology to BCI in a fully-implantable, wireless form factor, we had the potential to do for BCI what broadband modems did for the internet — completely transform the scope and scale of practical applications.
Explain the impact that brain computer interfaces could have for people suffering from paralysis.
The potential of BCI is much larger than paralysis, but the first people to benefit from it will be people with communication disorders caused by paralysis. There are about 150,000 people in the United States who cannot communicate by either speech or typing due to severe paralysis.
Their minds and senses are perfectly intact, but they are trapped in their bodies. Some people are able to move certain parts of their bodies in order to control assistive communication devices. Some people can move their eyes to use gaze-tracking “keyboards,” but some have even lost the ability to control their eyes and are now “locked in.”
Now, imagine for a moment that you are totally immobile and only have access to one means of communication to express physical needs, talk to loved ones, and engage with the outside world.
What if you could only type messages using a cursor and clicking on letters, akin to the frustrating process of entering movie titles with your remote control? How would this limit the way you communicate with your friends and family? Now imagine you had a technology that would let you virtually hand write messages, or even speak, just by thinking about what you want to say.
If you take a moment to reflect on your own experiences writing, typing, and speaking and then imagine being unable to move, relying completely on others for your basic needs, you can understand that a seamless communication device like this would be life-changing.
Is it possible for a BCI to provide this type of seamless communication? Or What kind of performance could we expect from a high data-rate BCI?
Conventional assistive communication tools, like gaze-tracking, have communication rates of less than 10 words per minute [wpm]. Wearable EEG technology, which can measure electrical signals in the brain from scalp electrodes, have similar communication rates. This is about as slow as typing into the Netflix app using your remote control.
Another interesting technology, Synchron’s “Stentrode”, is a stent-based electrode that is inserted through the blood vessels and has a communication rate of < 5 wpm in combination with gaze tracking. One appeal of this technology is that it does not require open brain surgery, however, at 5 wpm, the Pledge of Allegiance would take about 6 minutes to type.
The state of the art in implantable BCI today, the Utah Array by Blackrock Neurotech, has enabled a communication rate of 18 wpm by interpreting imagined handwriting. This is about as fast as an able-bodied writer. That work, which was done at Stanford University, is utterly inspiring.
We are building a platform with eight times more data than the system used in those “handwriting” lab studies, packaged for long-term use in a person’s everyday life. We anticipate our communication rates to be in the range of real-time human speech.
Why is this the right time for brain-computer interfaces?
In my opinion, the right time for BCI was probably 10 years ago, but unfortunately the science and technology to make it work as a commercial product was spread out across a lot of different fields, and the integration work to bring it all together was beyond what could be done in academia. Medical device companies like Medtronic are simply not set up internally to do this type of innovation, and until very recently, it has been an up-hill battle to get venture funding for BCI companies. There are two misconceptions that caused headwinds for BCI historically:
- Investors saw BCI as “deep science”; whereas in many of the first application areas it is really “deep engineering,” with strong existence proofs in the clinic.
- The lackluster performance of medical device as a startup sector, especially compared to software, had dampened enthusiasm for investors who think in patterns rather than fundamentals.
This is one of the reasons why VC funds like Prime Movers Lab can be so effective at accelerating progress in fields–like BCI–where there is a backlog of science and engineering, stalled beyond a modest energy barrier and waiting to unleash a flood of value creation.
Recent Pew Research showed that people are interested in brain-computer interfaces for medical purposes but not as much for commercial applications. Can you explain why medical applications are gaining more acceptance?
That poll is consistent with my experience speaking with people about BCI. I have never met anyone who didn’t want to develop therapies for people who are paralyzed. Using BCI to help people who are paralyzed, blind, or suffering from a severe neuropsychiatric illness that does not respond to medication is pretty uncontroversial.
In my experience, most people also see value for using BCI for mood disorders, with all of the same caveats that one would raise when thinking about traditional pharmaceuticals or psychedelics. That is, what exactly do we define as a mood disorder, what should “baseline” look like, and is it okay sometimes to feel sad?
I think opinions on BCI are a classic example of “most reasonable people agree” and “the extreme opinions get the most airtime.” A small number of BCI alarmists write click-bait articles, and a few healthy people claim they would sign up tomorrow for a BCI just to play video games. However, most people see BCI for what it is now–and will likely be for the next decade–a powerful medical technology.
What makes Paradromics different compared to Neuralink and other players in the space?
Paradromics and Neuralink sit apart from the rest of BCI companies in terms of capabilities. We are the only two companies building high data-rate devices, meaning recording single-neuron signals from 1,000-plus electrodes. Given the several-year lead we have versus other companies, we will likely remain the two major players building “broadband” modems for the brain for the foreseeable future.
Paradromics and Neuralink both have excellent engineers and scientists, and I think we are both building the right systems according to our own assumptions.
Neuralink thinks the future of BCI is the replacement for the cell phone. Everyone will get them, and they will want to replace them about every two years. Accordingly, they spent a lot of time and money building a robot to replace neurosurgeons. They are aiming to lower the energy barrier for implantation (and replacement), so it can become a routine outpatient procedure. They haven’t prioritized device longevity because who wants a 10-year-old cellphone?
Paradromics thinks the future of BCI is transforming unmet medical needs in brain and neurological health into solvable technical problems. That makes sense not only in terms of human impact but also commercially: There is only so much people will (or can) pay for an iPhone, but life changing medical devices can be reimbursed at tens or even hundreds of thousands of dollars per treatment. We think patients will want the option of keeping their therapeutic devices for many years, and so we’re building our device to last. We don’t see current neurosurgical practice as a limiting factor for getting therapies to patients. In fact, since the surgical techniques are well established and currently very safe, we aim to be very conservative in our clinical procedures. That will have the effect of making fewer regulatory hurdles (and potential safety issues) for the first patients. We want to minimize the resistance to adoption within the existing healthcare system.
You host a great neurotech podcast. Can you share who your upcoming guests are?
We debuted season two of our Neurotech Pub podcast this spring, and like in our previous season, we are taking deep dives with neurotech subject-area experts. Our season premier featured an insiders’ discussion with three founders of neurotech startups: Brian Pepin of Rune Labs, Kunal Ghosh of Inscopix, and Carolina Aguilar of Inbrain Neuroelectronics. Stay tuned for upcoming episodes that will feature some of the very first patients to participate in cortical BCI implant studies. This will be a must-listen for anyone currently developing the BCI devices or investing in the space.
Who inspires you?
Three people who inspire me as a human being are Abraham Lincoln, Leo Tolstoy, and Fred Rogers.
I’m also a big fan of Frank Lloyd Wright’s architecture (thanks to my father), and if you’ve never visited one of his buildings, I highly recommend it. Architecture is a great form of non-invasive brain state modulation. Unfortunately, it’s a rather expensive one.
Three neurotechnologists who inspire me are Ed Boyden, Krishna Shenoy, and Winfried Denk. One pleasure of being a science nerd is that you actually get to meet your heroes.
My inspiration as a podcast host is definitely Melvyn Bragg, host of “In Our Time” on BBC Radio 4. I stole the multi-guest format for Neurotech Pub from “In Our Time,” although the pub aspect is definitely our own!
Where do you see brain-computer interfaces heading in the next 5 years? 10 years?
Ten years from now, BCI will be as obvious as a heart pump. What does a heart do? It pumps blood. What if someone’s heart can’t pump anymore, but we can’t find them a transplant, and we can’t grow them a new heart? Simple: We give them a pump.
What does a brain do? It takes in information, processes information, and outputs information. What happens if any of those processes break down due to a biology problem that isn’t directly fixable. Of course: We need a data interface with the brain! How else could we restore the data flow required for sensation, movement, or communication? When we look back at our current approach to mental health–treating network-level brain activity disorders with small molecules (pharmaceuticals) and using only behavioral feedback–we will likely be surprised it took so long to adopt technology-based solutions.
The previous decades have seen a “doubling” time for increased brain computer interface bandwidth of seven years, much slower than Moore’s law. The next decade is likely to be closer to doubling every two years, which will result in an incredible explosion of capabilities and applications. In retrospect, however, I think it will be the paradigm shift of considering the brain in terms of neural activity and data, rather than specific technical developments, that will mark the biggest advancements in neurological healthcare.
On a lighter note, what is your favorite BBQ place in Austin?
I’m afraid to reveal this publicly because currently it has Franklin’s-level brisket and no wait time, but if you are willing to keep a secret, I will disclose to you that it is Brown’s Bar-B-Que on South Lamar.
Prime Movers Lab invests in breakthrough scientific startups founded by Prime Movers, the inventors who transform billions of lives. We invest in seed-stage companies reinventing energy, transportation, infrastructure, manufacturing, human augmentation and agriculture.
Sign up here if you are not already subscribed to our blog.
