Mike Yu Chi, right, and Tim Mullen are the lead researchers on the study. Both are UC San Diego alumni. Chi cofounded Cognionics, which developed the EEG headset featured in the study. Photo: UC San DiegoEvery Thursday, Laboratory Equipment features a Scientist of the Week, chosen from the science industry’s latest headlines. This week’s scientist is Mike Chi, who led a research team in developing the first portable, wearable EEG headset.

Q: What are the benefits of having the device be portable that you wouldn’t have when confined to a lab setting?
A: The biggest advantage is that it enables researchers to study neurophysiological responses in real-world as well as resource constrained (e.g., limited time, infrastructure) environments. It's really enabled by both the portability and the fact that the electrodes are dry.

Q: What exactly is it about the dry sensors that makes the device better suited for real-world applications?
A: Wet electrodes require time-consuming preparation, anywhere from 10 to 40 minutes for each subject. In addition, they leave a messy residue which takes time to clean up. In comparison, dry electrode headsets can be applied in two to five minutes and require zero cleanup. Dry electrodes furthermore do not require consumables for each use.

Combined with a portable headset, this means that researchers can take the system to almost any setting and quickly gather data. The fast application of the dry system also potentially enables larger scale studies (e.g., recordings on many people at the same time). Perhaps most importantly, the simplicity of a dry system allows for subjects to take the system and do recordings at home without needing technician assistance.

Q: What impact can this have on the general public and/or future research?
A: Currently, we're pursuing two tracks that may have long-term benefits for research and the general public.

The first is helping researchers do recordings in environments that were not previously possible due to the size and bulk of traditional EEG devices. Here's a few things that some our collaborators are working on with our systems that exemplify tasks that were previously difficult or impossible.

1. Taking EEG recordings in an airplane to monitor pilot cognitive workload.

2. Doing EEG recordings on a group of children in a classroom to understand how they respond to music lessons.

The second is enabling new medical applications where having a dedicated EEG technician and the time it takes to setup a recording is not practical:

1. Quantifying the severity of stroke in the ER

2. Detecting epileptic seizures in a cost and resource constrained clinic

Q: What sort of issues do you run into when attempting to gather data while a subject is moving/exercising?
A: Movement is very hard for any EEG system. There are numerous external artifacts due to the movement of the sensors, with respect to the skin, sweat, movement of the cables - all of which can generate noise larger than EEG. During movement, a subject will also generate other physiological signals (e.g., muscle activity) that can also contaminate the EEG. This is a long-standing problem that affects every system and experimental protocol.

Dry systems will have a harder time coping with many of these artifacts than wet due to their inherently poorer coupling to the skin. One of the things that we've really spent a lot of time on is carefully optimizing the sensors, the mechanics and the electronics so that they all work together - which is critical to achieving good signal quality. I think we're one of the few groups that can get all three aspects working well and our systems can achieve good signals under most light ambulatory conditions.

Long-term, I think the solution is improved system designs as well as new software techniques, such as those developed by our partners at Qusp, to clean contaminated EEG data have made significant strides in the past few years.

Q: What are some of the next immediate steps for further development of the headset?
A: We are always working on improving the system. Right now our system works quite well for light ambulation (e.g., walking) but heaving motion (e.g., running, jumping) is still a challenge. We're working on improved electro-mechanical designs to make the system more resilient against severe motion artifacts. We're also constantly working on increasing the number of channels our headsets support while keeping the system easy and fast to apply.