A type of assistive technology, called brain computer interface (BCI), has been around for years with much room for improvement. We sat down with the 2016 ALS Assistive Technology Challenge winner, Dr. Peter Desain from Donders Institute for Brain, Cognition and Behavior in Nijmegen, The Netherlands, who invented NoiseTag BCI that gives a novel spin on BCI technology. It is faster, more comfortable and efficient and easier to use than ever before. Learn more in our interview with Dr. Desain.
What is NoiseTag BCI?
Traditional BCI works OK but is quite slow. Our NoiseTag BCI is a breakthrough! We changed it so that all the buttons on the screen flash in a pattern (see images below). It is almost like bar codes read out in time or even like a Morse code signal. The user selectively attaches to one of the tags and in response, the BCI can see your selection. It then outputs the letter or button of the keyboard you are looking at. We call it a NoiseTag because these patterns look like noise. They are less annoying than periodic stimulation. It is like looking at the flicker of an old movie.
How quickly can a person learn the BCI?
It does not take much of an effort. Just look at a specific button or key. The old BCIs need to learn the user responses during an essential calibration time. We found a method that turns BCI into plug and play. The first button you look at will take the system a bit longer to figure out, taking about 30 seconds. Then the second button goes down to 10 seconds. Then the 3rd-4th is down to 1-2 seconds. A person can get up to 1 button per second.
What is unique about your system?
Right now, we use a full cap with wet sponges. That works very well and is comfortable, but we are moving to a headband that is better. The headband uses dry electrodes, so we do not have to use water. We collaborate with a company called Cognionics to develop headbands for ALS patients who have a headrest. We will design the headband so that the headrest will not move the electrodes. We want to make it as comfortable as possible. Because even a small pressure on the head from the electrodes after a while is not very pleasant.
What sets you apart from other BCI technology? Other than it will be dry, fast, comfortable, and using a headband instead of a cap.
Usually BCI is very brittle and easy to upset when one electrode gets loose. Ours is adaptive to electrode contacts. We calibrate on the fly with each trial and so when an electrode becomes bad, it is just left out of the calculation. It is much more adaptive.
Also, another type of adaptivity is really nice. The computer keeps collecting data until it is confident enough in which letter you are looking at. So if you are really concentrated, it becomes really fast. But if you are a bit tired, it has the same accuracy, but at a slower pace. You can set the desired accuracy and it waits and knows that you reached that level. So for patients that hate to make mistakes, they can set it very high. Others do not mind using the backspace key, so they set it a bit lower.
The last thing that makes it stand apart from eye tracking is that can be attached to a tangible object, like a door knob. It is not restricted to a computer screen.
So you can control a door knob by putting the light NoiseTag on the door knob?
Yes, then when you look at the door, the door will open. It is a very natural mapping. You do not have to go into your program and find via menu that you want to control the door. Just looking at it can make it do the action. You could in theory put the NoiseTag on anything you want to control. (See the diagram below.)
How did you get involved in BCI for ALS?
It was a funny road. I did a lot of research on music cognition and perception and started measuring the brain response. I did work with imagined music, like when you have a song in your head without openly singing it or hearing it. I was able to find which rhythm you were imagining. Then I suddenly thought, wow we could use this to help people who are completely paralyzed to communicate. That made me move to the BCI field. That was the beginning. I do not have a proper date, because that was a long time ago. At first it worked nicely, but I and many people became frustrated about how slow the progress was. A few percentage points every year in P300 spellers, another type of BCI technology output. We were very happy when we suddenly found that our new method is an order magnitude faster and much more reliable. Suddenly it makes things possible! We are so happy to receive the prize, so that we can build a product that people can actually use and use much more easily.
How did you start working with Evy Reviers from the Belgium ALS Foundation (The ALS Association counterpart in Belgium)?
We applied for European grants for funding together and it turned out to be a very nice collaboration. It is great to have such a partner to get information to patients and to help us validate our system through testing with patients.
When are you going to start testing your BCI technology with people living with ALS?
We are not yet, but we will and looking forward to that. Right now, we are working with healthy people and will work with Evy in her patient population. The Belgium ALS Foundation is very active with patients and getting what they need and setting up exchanges. They are a very valuable partner to us. We will also work with our hospital here.
Do you have a message for people living with ALS?
I met so many ALS patients at the ALS Assistive Technology finale event in Dublin. Before, we only did one test in the lab with a patient, just before we went to Dublin, in order to prove that a patient can use it. Then I met some of the judges and people that participated in our demonstrations. I am really impressed how they want to help their own community to have better access to support communication.
Previously, I tried to setup a collaboration with an industry, but often they object since the ALS market is too small. That is an awful approach to the situation where people, that close to the end of their lives cannot communicate. It is so important that they can still express love to the people around them. Even frustration or other types of emotions. Even if it were for one patient, it would be worthwhile to create a better communication.
What are you plans with the award money? Congratulations!
We set up this consortium of collaborators and we plan to use the award to make NoiseTag BCI available to the market. We will connect to a few existing communication programs, since they are already out there. And we will create an open standard, so that a developer of assistive technology can easily connect to the BCI. We will make it really open, so that they can use it. We will also develop the hardware for headset and make the app available. It will be a concentrated effort in collaboration with these partners. I think the time is right now. The method is good enough for a great end product.
What is your timeline?
Two years. Early prototypes will go out earlier. We will quickly start with collecting feedback from patients from our BCI tests. One important issue is how well it works for patients that cannot use eye tracker anymore. Eye trackers are only a good solution for specific period of time. It is important to find out in which patients our BCI still works, when eye trackers do not work anymore.
Is there anything else you would like to add?
I really like that many different organizations are coming together to organize this prize. I think it is worthwhile to join efforts and share what comes out.
Donders team members:
- Peter Desain, Donders Center for Cognition (NL), Project Leader
- Joost Raaphorst, Jan Groothuis, Janneke Weikamp, Radboud University Medical Center (NL), Neurology and Rehabilitation Departments
- Bart van de Warrenburg, Donders Institute for Brain, Cognition and Behavior (NL), Knowledge Utilisation
- Evy Reviers, ALS Liga (BE), Patient Associations, Communication
- Mike Chi, Cognionics (VS), EEG equipment
- Merijn Klarenbeek, WeBoost, (NL) Project Management
- Peter Ossenkoppele, rdgKompagne, (NL) Assistive Technology Provider
We congratulate Dr. Desain and his team and cannot wait to hear more from them in the future as they develop towards commercializing this innovative new BCI technology! We will keep the ALS community updated.
To read more about the ALS Assistive Technology Challenge, please visit http://www.alsa.org/research/als-assistive-technology-challenge.html.
To learn more about assistive technology, please visit http://www.alsa.org/research/focus-areas/assistive-technology.
Read about the other AT Challenge winner – the Pison Technology team here.