Brain-computer interfaces are all the rage in medtech, with practically each week bringing the debut of a new “mind-reading” headset or neuron-stimulating implant.
The notoriously wary FDA has even joined the crusade. In the last year alone, it offered up a leapfrog guidance specific to the technology, doled out a handful of breakthrough-device designations to brain-computer interfaces and, in April, issued its first approval to a system that translates brain signals into movements of a hand and wrist brace, allowing stroke patients to relearn how to grasp objects.
And while these rapid-fire developments make it easy to get swept up in the hype surrounding brain-computer interfaces, because of how sensitive neurotech can be, it’s important to realize that plenty of it is just that: unsubstantiated hype.
“More often than not, there’s a lot of people promising things that the tech can’t actually do,” Justin Sanchez, Ph.D., the technical fellow in charge of Battelle’s life sciences research business, told Fierce Medtech.
“It’s so important to get excited by the potential of neurotech and where it could go, but also to be really grounded on that technology and what it does and what science needs to happen in order to make it a reality,” he continued. “Understanding the difference between what is real and what is not when it comes to neurotech is such an important issue, and it’s one that, at Battelle, we take extremely seriously.”
Battelle’s brain-computer interface work began in the realm of spinal cord injuries with the development of a system that analyzes brain activity to detect a paralyzed individual’s intent to move their hand or arm, then uses that information to stimulate those muscles through a forearm sleeve.
That fundamental technology has since been expanded into stroke rehabilitation and, through Battelle’s long-running partnership with the U.S. Defense Advanced Research Projects Agency—where Sanchez was previously director of the biological technologies office—the ongoing development of a noninvasive brain-computer interface that involves an external headset and injected nanoparticles for neural stimulation.
In a Q&A with Fierce Medtech, Sanchez offered up the two major questions that can separate science from sci-fi in brain-computer interfaces and plotted out the technology’s next moves.
Responses have been lightly edited for length and clarity.
Fierce Medtech: To start, can you talk about Battelle’s work with brain-computer interfaces so far?
Justin Sanchez: For a number of years, going back to 2013, 2014, we have developed neural interfaces to both the brain and the body that can detect when the brain and nerves and muscles are becoming active. We’ve got algorithms that can interpret that information in real time, and we can use it to do remarkable things like reanimate a limb after a person has been paralyzed—or we can use it even beyond the medical sense, in sports applications or entertainment applications.
We’re able to do all of that because we do exquisitely know how the tech works. We know how to interface it with the body, and then we can design it in a way so that it delivers the highest level of benefit to the people who are actually using it.
FM: Brain-computer interfaces clearly have a lot of potential in medicine and are really buzzy right now, but what would you say is the key to differentiating between “the real deal” and what may just be empty hype?
JS: One of the first red flags is if somebody makes a promise that’s totally science-fiction or too good to be true. That’s your very first sign to think, ‘Wait a minute, what can people do today versus what might be many years in the future?’ You should always first evaluate that: Is this in the realm of possibility, or is this a fantasy?
The second question should be: What can the technology actually do to achieve whatever that grand vision is? Let’s say you want to have full cognitive control over a machine using just your brain activity. Deep understanding about the subnetworks of the brain that are involved in that is the first bit of knowledge that one needs to have. The second part is that there has to be technology that can get to those subnetworks of the brain with sufficient resolution so that an algorithm can pull that information out and actually make sense of it. And then the third part is, even if you can interpret that information in real time, are you doing something useful with it?
You can go on Amazon and buy an EEG headset—a noninvasive headset that you just place on top of your scalp—but that technology is fundamentally incapable, through the biophysics of it, of getting to all of the brain’s networks with sufficient resolution to achieve the level of fidelity that one would need to control cognitive function. So, if somebody were to come to me and say, ‘I’ve got this noninvasive EEG headset that’ll allow you to upload your brain to a computer,’ I’d be like, ‘Wow, that sounds interesting, but the technology really can’t do that,’ and I’d know that what they are saying is not true.
FM: Even among those brain-computer interfaces that are grounded in legitimate science, only a few have moved into clinical trials, and just one has been approved by the FDA. Where’s the roadblock? What’s keeping the market from being flooded with these devices?
JS: There are a lot of different devices that are being used in humans under investigational device exemptions. These are prototype devices that are being used in research studies, and the information that is garnered from those research studies will eventually be used to help get those devices through the follow-on regulatory pathway so they can be cleared at some point to be prescribed by a physician.
With that being said, while those technologies are maturing, they’re not at the point where they can be brought to market in a sustainable way so that they can be prescribed and produced and manufactured and so that companies can do this over and over again. There is a huge gap right now between the promise of neurotech and the productization of neurotech.
That’s an element that we understand very deeply at Battelle, and it’s also one that we are going at in a number of ways. There might be variants of neurotech that are more accessible today by broader masses of people that can be made into a real product that gets to the market quickly, and that can be happening at the same time that one is pursuing a regulatory pathway that is maybe many years in the future.
So, back to your original point: There is a lot of neurotech that’s maturing, but there isn’t yet neurotech for the masses because there aren’t necessarily the markets yet, and there aren’t commercialization pathways yet that have been done well. That’s something that we’re working on at Battelle: getting the real technology that’s grounded in solid science and provides a benefit to users, and then coupling that with the productization cycle that makes sense for the development trajectory of neurotech moving into the future.
FM: Once those productization cycles start moving, what untapped areas of medicine do you think we’ll see brain-computer interfaces starting to address?
JS: This is where medicine and some other areas merge together. First off, to get real neurotechnology utilized in medicine in a big way, it’s about addressing patient populations in which there’s a lot of people that need something new, where current practice is not delivering what the patients need. Stroke is a great example of that: There are many, many, many people living with stroke who need rehabilitation, and a technology like neurotech could make a huge impact—there’s a market and a demand for that.
The same holds true for mental health. That’s a domain that’s primarily treated right now with drugs and pharmaco approaches, but those are not specific and don’t always treat the underlying issues. That’s a place where neurotech might be able to not only diagnose but also treat individuals in ways that the existing standard of care can’t do. And again, there’s a very, very big patient population that can benefit from neurotech.
Here’s where it ties into some of the broader markets: In areas like stroke rehabilitation, there are new, emerging technologies that are using gaming as part of the rehabilitation process that could provide more interesting rehabilitation regimens that neurotech is very well aligned to being coupled with. We’ve started exploring some of those techniques—using virtual environments and experiences as part of telemedicine and neurotech—and as some of those neurotechnologies mature in the medical domain, there are also likely pure gaming uses for the neurotech that’ll provide more immersive experiences that the gaming community would very much desire. That immersion that helps somebody who’s living with stroke and going through rehab will also help people who are trying to have a really fun gaming experience.
The duality of neurotech—that it can be used in both those spaces—is also what sets the conditions for getting neurotech from where it is today to becoming something that you or I or anybody might ultimately use. It’s groups that truly understand the fundamental science of how it’s applied and where it can be used that are ultimately going to succeed.