Building BCI For the Real World: A Chat With Dr. Benjamin Rapoport of Precision Neuroscience
Clinical Practice Guidelines, Explantation, Micro-ECoG, R&D Strategy, and More
NEUROTECH FUTURES recently sat down with Dr. Benjamin Rapoport to learn about his work as co-founder of Precision Neuroscience and his perspective on several emergent topics in BCI commercialization.
By Naveen Rao and Michael Nolan
Benjamin Rapoport, Chief Science Officer, Co-Founder – Dr. Rapoport is a neurosurgeon specializing in minimally invasive surgery and a co-founder of Precision Neuroscience, where he oversees the design and development of the company’s products and surgical implantation techniques. He has been developing brain–computer interface technology for nearly two decades. He has an MD from Harvard Medical School, a PhD in electrical engineering and computer science from MIT, and was on the eight-member co-founding team of Neuralink.
Hi Dr. Rapoport. Why did you co-found Precision Neuroscience?
The short version is that by 2020, when (co-founder and CEO Michael Mager) and I got together and formed Precision Neuroscience as the company it is today, we had already learned a huge amount about what worked well. We had learned a lot about what was possible, and what we felt was necessary to bring brain computer interfaces from scientific possibility to medical reality.
You know, you don't start a company to do new science. You start a company to to do things that can't be done in an academic setting. You have to know a technology is ready for that, or that there are things that can only be done outside of an academic setting with what some people now call “deep tech.”
We felt as early as 2015-2016 that the science problems in brain computer interfaces had effectively been solved, and that the implementation and translation really needed to be addressed in a commercial setting. And by that point Michael and I had already been talking about forming a BCI company. I think a lot of people converged on the sense that it was time to to put the commercial resources and professional engineering resources and other sorts of things outside of the academic setting behind BCI, to bring potential into clinical reality.
You may know that I was part of the Neuralink founding team. I left in 2018 with the firm conviction that I had going in, which solidified after two years, that in order for BCI to become not just science but clinical reality, it needed to be extremely safe and scalable. Some changes were needed in the way the technology was being implemented, in terms of the nature of the electrodes and the implantation techniques being appropriate for research grade science, but not scalable to clinical implementation.
We felt that the electrodes themselves had to be non-traumatic to the brain, while still being able to record very high resolution signals from many electrodes to generate the kind of high resolution data required to decode brain activity and intention. We felt that the surgical procedure also needed to be minimally invasive and low risk, low morbidity, just like many other procedures that have come to scale and define the medical technology landscape over the years.
Implanted cardiac devices and the last 20 years of neuromodulation devices have really learned those lessons. BCI, as they were being brought out of science into the initial startup world, hadn't fixed that form factor or some of those engineering details. And so we started Precision with the conviction that brain computer interfaces needed to be scalable, safe, and minimally invasive.
You chair the Clinical Practice Guidelines workgroup for FDA’s implantable Brain Computer Interface Collaborative Community (iBCI-CC).
What do clinical practice guidelines for BCI encompass? How do you approach this development work in such a nascent clinical field?
I'm going to try to answer that in a couple of different ways. First, it always ends up that when you actually try to get something to work in the real world, no matter how simple it may seem, there are little wrinkles that end up being nuanced and complicated.
So, clinical practice guidelines are meant to help the field as a whole learn from the past and implement all of the details of something highly complicated for real human beings. We’re developing treatments for real human patients in a way that adheres to and respects the experience of the past, and the data that's been gathered in the past.
They're meant to be living guidelines, adaptive and responsive to the experience of the field that they belong to. It's interesting that in a field that is so young, where there isn't actually a lot of clinical evidence, that we're thinking about clinical practice guidelines from the beginning, but I think it's really important. And so we're going to try to use whatever evidence we have from the very beginning to encourage the field to to move forward in a data driven, evidence-based way.
Collectively, there's been a few dozen long term implanted BCI patients in small clinical trials over the last 20 years and there's a number of experts from various disciplines, clinical and scientific and other areas who've been involved in that. At this point, we want to understand the landscape of experience and try to represent that in a kind of consensus for the patient journey, from evaluation of candidacy for a brain computer interface through the implantation and training and early experience into the medium to long term experience with an interface.
And the goal of the FDA’s workgroup is to iterate on this. We're in a period of time now where we're looking at a number of relatively advanced technologies starting to design early feasibility and pivotal clinical trials. And I think over the next three to five years, we'll see a big jump not just in the available data, but the amount of clinical evidence and clinical experience that the field as a whole has with BCI patients, and we really want to be ready to capture and analyze and reflect that experience back into the community.
Building on that, let’s talk about a topic nobody in BCI talks about: Explantation.
People are going to use BCI for years, but at some point, potentially replace them or have them taken out. What is the clinical or operational strategy for doing this safely, in partnership with patients?
The surface array technology at Precision was really designed, in part, with that in mind. Not because we want electrodes to be explanted. But I think that if you had asked many people 15 to 20 years ago, there was this idea that an implant was supposed to be designed for the lifetime of the patient. And so this idea of explantation, was supposed to be “not thought about,” and that all of the design parameters were put in to be able to last 100 years, or something like that, right?
And I think, increasingly, the field realizes that that is a very tall order. Not only is it really difficult for any foreign body to survive in the in the body for a lifetime. Beyond the biological issue, it’s also that technology changes over that period of time. And beyond the need to remove them, somebody might want to be able to upgrade at the hardware level.
Certainly, explantation has been done with many of the early Utah Arrays, among BrainGate patients. There have been many examples. It's harder when you have a technology where you've basically made holes in the brain and there's scar tissue around an implant. Ian Burkhart is an example of a person who had a safe explantation. There are quite a few others. But, you know, it’s not ideal if the device is not designed with that in mind.
And so one of the advantages of a surface array technology is that, because it doesn't cut into the brain, it doesn't, it doesn't form a glial scar, you know, in the brain tissue. There's a long history of surface electrodes, both in the brain and spinal cord, being explanted and replaced, upgraded and changed safely. There's always some scar formation that forms wherever an incision is made. But in the case of a surface array, the tissue incision is not in the brain, and it's not at the site where the electrodes touch the brain surface.
And so we try to avoid scar formation and adhesions such that when the electrode is removed, the brain doesn't get damaged. And we have some medium term experience in our animal studies of safe and uneventful explantation. Of course, this kind of thing requires data. So we don't yet have the the long term data that we would need to show that we can do that in a human, but certainly we've been thinking about that.
What kind of referral pathways inform the patient’s journey to getting BCI in the near future? It’ll still be brain surgery, even if it’s less invasive. How will patients, caregivers, medical teams make shared decisions?
People find out about therapeutic options in all kinds of ways, right? It used to be much more common that you needed to find out about treatment options from a doctor. But, you know, in today's world, people find out in all kinds of ways, from friends and family, from the internet, from their doctor. So, I think the answer to that question is that it's going to evolve over time.
Our first experience with this is really in the early human clinical trials that we're doing now. The ones that we're looking towards for permanent implants in terms of early feasibility studies in the pivotal trial, in those the referrals come through. The physicians that are at partnering sites. Those may be surgeons, neurologists, rehabilitation doctors and others who are close to the BCI field. And also people like myself are taking care of patients and usually our practices are engaged with patients and families who are impacted by the diseases that we're trying to address.
And so there is a kind of grassroots element to referral networks, as there always has been at the clinical trial level. In addition to that, the public consciousness around brain computer interfaces now certainly helps. Synchron and Neuralink have done something nice, which I think most of us are going to do a version of, which is to have essentially a registry of interested parties. That allows patients to be considered in a proactive manner. So I think there's lots of ways that interested and eligible people may enter the funnel. It’s going to continue to evolve over time.
What specific applications are you working on, in terms of clinical indications, patient populations, and use cases?
All the BCI companies right now in the early clinical trial phase are focused on restoring function to paralyzed people. And exactly what that means, to be very rigorous, is not totally defined from a clinical trial outcome standpoint. But at an intuitive level, we and others are trying to restore the ability to interact with the digital world, for people who can’t use their hands. To give you more detail - A quadriplegic person who can’t use their hands should be able to have thought-based control of a standard OS, with as fluid or intuitive interaction as you or I can. Hopefully, even better than that.
So an example of a target patient right now is…imagine somebody like you or I, who has a regular desk job, who gets into a terrible car accident, and was previously supporting their family, and overnight becomes totally dependent on others for their care. They just can’t physically hold a job or earn a living or talk to loved ones without assistance, 24/7. But, they are still cognitively intact, lucid, So, the intent here would be to restore the ability of that person to have access to the digital world, in the same way that we do now, without hand or voice activation, to do it with thought control.
We know that that’s possible, so the goal is to make that seamless and high functioning enough that it crosses a high feasibility and quality of life threshold for individuals who’d want to use it. Moving a mouse on a computer screen, gestural control, click, and type. There are many things that will come after that, but that’s the bar that we need to pass now.
The full conversation, which continues below, is for paying members only.
Here’s what we cover:
Differences between standard ECoG and Precision’s Layer 7 Array
Recording vs stimulation, at Precision and in neurotech/BCI more broadly
Why Precision’s World Record for implanted electrodes matters for patients
Calibration for stability in different types of BCI
Inside Precision’s R&D partnership with Lawrence Livermore National Lab
Device longevity and target device lifespan for BCI
Insights from 16 completed investigational implants, going into FDA trials
To continue reading, please consider upgrading (group discounts available) or try our new referral program. Hit reply with any questions!