Brain-computer interfaces are still years, and several FDA approvals, away from being available on the market. Even though industry leaders tout their eventual use for the general public, the first users of these technologies have been and will continue to be people with disabilities.
One of the primary uses for BCIs is to provide better communication for people who have been paralyzed. Paralysis affects over 5 million people in the United States, which includes diseases like ALS, in which neurodegeneration complicates communication as the disease progresses.
STAT spoke with Mariska Vansteensel, a neuroscientist at University Medical Centre Utrecht in the Netherlands and president of the international BCI Society, about the field and about a new study that she and her colleagues just published in the New England Journal of Medicine.
Vansteensel has spent most of the last decade experimenting with BCI at-home use, rather than in a laboratory, and her team chronicled an ALS patient’s use of the device for over seven years. The 58-year-old woman was diagnosed in 2008 and obtained the device in 2015. Her BCI usage increased as time passed and her paralysis increased. The technology could prove fruitful for ALS patients as their eye-gaze devices lose utility as eye muscle control wanes.
This interview has been edited for length and clarity.
When you were planning this study in the early 2010s, what were you hoping to achieve by following this person for almost a decade?
At that time, no one with motor impairment had ever tried to use a BCI implant at their home. So we wanted to get data from independent, unsupervised home use from someone who is severely motor impaired. Everything was within laboratories, and for the most part that really hasn’t changed that much, most of the research is still in the lab.
We need to learn what happens in settings of daily living. Is it usable? What are the factors that play a role? What are you going to encounter? What things that you have never thought about. Answering those questions is important, if we want to clinically implement this technology.
Why did the study participant increasingly rely upon the BCI to communicate over time and then quickly end their use?
What we have seen in our participants is, people with ALS choose the most efficient way to communicate their messages. That could be just blinking in response to yes-no questions. It could be a speech BCI, where software tries to detect what words or sentences someone wants to say. We have had success with a click-based signal, in which the patient “clicked” on a letter or target by moving their hand to stimulate neural activity. The simpler approach can be useful in situations where alternatives are not working.
It is unlikely that the hardware is to blame. At the end of the study, she was unable to voluntarily produce any signal changes. We got a variation in the signal when we brushed her hand, which means that if the signal changes would have been there we would have detected them. But she was unable to produce the signal changes herself. The neurons needed to voluntarily produce those signals have just been degraded.
What advantages does a click-based BCI hold over other forms of BCI and other communication methods such as eye-tracking devices?
They’re more helpful in late stages of ALS and in certain circumstances where regular technology is not working. For example, when she went outdoors in the initial years of her using the BCI, the eye gaze device was not working well because of the variable lighting. So it’s a situational use, but it also relates to disease progression.
A click-based BCI is particularly useful in more severe circumstances. A very fast BCI decoding in those late stages is going to be very, very complex. We want to take a safer route and take smaller steps and see what’s feasible in this very late stage.
I think speech BCIs may be useful for earlier stages of the disease, perhaps even before the eye-gaze device, but I have yet to see whether speech BCI is useful, usable, and able to produce the correct signal changes in the late stages of the disease.
How does the BCI technology used in this study differ from other BCI tech that’s out there?
We have used subdural, electrocorticographic electrodes. From our study, it can be concluded that non-penetrating electrodes offer a high-quality signal for many years. There is also a large portion of BCI research conducted with intracortical, penetrating electrodes. Because of the stable long-term performance we report in our study, we believe that non-penetrating electrodes are more likely to be sustainable in therapeutic applications of BCIs.
To ensure that complications of BCI implantation are limited, future clinical implementation of implanted BCIs ideally involves approaches that are minimally invasive, requiring small burr holes in the skull for example (like we used in this study), rather than large craniotomies.
How does this observational study change our idea of the later stages of ALS?
I don’t think this study changes our ideas about ALS as a disease. However, I do believe that our work provides novel insights into the actual usability of BCIs for people with ALS. We show that this technology can be highly valuable as a communication tool for many years, even despite the limited functionality it offered. In addition, our work shows that the usability of the BCI can be situational (e.g., indoors/outdoors), and evolves as the disease progresses. Finally, this report indicates that there are some limitations to the usability of BCIs for people with very late stages of ALS. Although more research is needed to verify this observation, I believe this report contributes to painting an honest, realistic picture of the value of BCI technology for this group of users, which will be important for those considering BCI implants when they become clinically available in the future.
What are your thoughts about where the technology is headed and if there are any research paths that you see as particularly promising?
My own research is highly focused on assessing the internal and external factors that play a role in the performance of BCIs, including time of day and sleep. I am also fascinated by the progress that is currently being made in the space of speech decoding: identifying from brain signals which words someone tries to speak. This concept makes implanted BCIs attractive also for people with lesser forms of motor impairments, such as those with earlier stages of ALS who start to have difficulty speaking out loud. Implanted BCIs may also become attractive for populations that historically have received less attention in the BCI field, such as children and adolescents with severe motor impairment, like cerebral palsy. We are currently investigating the feasibility of applying implanted BCIs in these groups.
Many BCI studies involve a tiny sample size, using devices on a small number of participants. How does this limited sample size influence the data?
If we want to understand if this is going to be a problem for every ALS patient, we need to pull together data from different institutes. I think, as a BCI community, and I think I can say it as the president of the BCI Society, we need to bring together data from different studies and from different teams. To be able to do this comparison, we need to start doing better in pooling data.
Why is it important for researchers to show that BCIs can work at home?
If you try to establish at home use of a BCI, you will need to have good user interfaces. You will need to have a very stable system. It needs to be manageable for the caretakers to actually switch on the system, connect the cables and whatever is needed. It forces you to think about user-friendly technologies.
The field is making huge leaps at the moment. The artificial intelligence, it’s really impressive. But the pipelines securing independent home use that families and caretakers can use to maintain the system, that also needs to be set up.
STAT’s coverage of disability issues is supported by grants from Robert Wood Johnson Foundation and The Commonwealth Fund. Our financial supporters are not involved in any decisions about our journalism.