EEG Football Helmet

Dec. 2021 - May 2022

Project Assigned

Under the Human Neuromechanics Research Lab at UF, five undergraduate engineers and I were assigned the task of creating and prototyping a football helmet with the capacity to capture human brainwaves via electroencephalogram (EEG) electrodes.

Context

The overarching question posed by our Principal Investigator was whether or not we could measure brain dynamics in a human while they played football. In other words, the idea was to see if it was possible to record brainwaves in football players while they played football in hopes of being able to use their brainwave information to diagnose potential concussions. These findings could ultimately help coaches on the sidelines determine whether or not their players have accrued serious damage during a game, at which point players could be closely monitored or benched.

Objective

Since we were only given one semester to tackle the issue, our semester’s objective was simplified to the following: to innovate a way to measure and record high-fidelity EEG signals via electrode implementation within a football helmet.

Assumptions

The patient/football player is bald; therefore, we do not have to design/buy electrodes that go through hair
Do not worry about football standards or industry standards; this product will not be commercialized in the near future

Additional Note

In other words, all we needed to do this semester was create a working prototype that could pick up clean signals consistently. Our PI just wanted to see if it was even possible to create an integrated EEG football helmet that could pick up clean EEG signals consistently before adding more restrictions/requirements. The idea was to simply this problem as much as possible and add complexity as certain milestones were achieved, ultimately resulting in a football helmet that could be put on any football player before a game.

Personal Contribution to Project at Each Phase

Progress Report #1

The first step for the group was to carry out a literature review on EEG signal collection and on all previous attempts at integrating EEG electrodes into a football helmet. Our lab mentor guided us by providing a few papers for us to skim through to gain familiarity with EEG signals and electrodes; however, we were also encouraged to search for papers ourselves, especially those dealing with previous EEG football helmet attempts. We created a Word document where we compiled all our background notes on EEG concepts, and we did the same for the football helmet attempts.

There were four notable attempts at integrating EEG electrodes into a football helmet in the past, which we divided and conquered. One of my teammates and I were responsible for researching the NoMo Helmet, created by Colombia University, and documenting reasons for why it may be failing, potential considerations to keep in mind when we begin creating our prototype, and general pros and cons attributed to the helmet. Once we completed our research as a group, we condensed all our notes into a progress report PowerPoint that we presented to our lab mentor and PI, which also included design ideas we had for our prototype. Below are a few slides from that presentation.

Progress Report #2

After completing the preliminary research, we divided into two groups: an electrodes team and an electronics team. The electrodes team was responsible for researching and purchasing electrodes, innovating unique ways to install the electrodes in the helmet whereas the electronics team was responsible for researching electronic solutions to enable EEG signals to be recorded from electrodes. I joined the electronics team; however, there was room to support the other team with brainstorming so long as it didn’t interfere with your assigned group’s main responsibilities.

An idea I shared with the electrodes group was to see if we could install the electrodes into the pads; in other words, cutting into them to ensure that the electrodes sat flush with the padding. This initial idea was expanded upon by the electrodes team and implemented for the entirety of the project, which can be seen below in an attempt to install gold cup electrodes into a football helmet we were provided with in the lab.

As the electrodes team continued with their research, the electronics team proceeded to research ways to pick up EEG signals, amplify them, and get them to display on some sort of software. Our lab mentor found an old Cyton PCB in the lab with some other hardware made by a company called OpenBCI, which I researched and discovered worked perfectly for our application. After researching how all the hardware functioned and downloading its respective software, we conducted tests on the gold cup electrodes to see how well they could pick up brainwaves. Two analog electrodes were placed on my forehead and the back of my head, respectively, to generate EEG signals while two other electrodes were placed on my earlobes to function as a reference and noise-cancelling electrode, respectively. The set up and results can be seen in the images below.

At this point, we had come up with a viable electronic solution for recording and displaying EEG signals and a means of installing the electrodes within the helmet (flush with its internal padding). The electrodes team also discovered another type of electrode that would work with our electronic solution and would be quicker to implement and install since the gold cup electrodes required a conductive gel. Once again, all this information was condensed and presented as our second progress report to our lab mentor and PI.

Progress Report #3

After purchasing the new electrodes, the electrodes team cut into the football helmet padding with a Dremel to create enough space to allow the electrodes and their wires to fit sit flush inside. The electrodes were installed, and the electronics were placed within a small pouch inside the helmet, which can be seen below as our first prototype.

However, after testing out the football helmet on a teammate and on myself, we realized that we were not able to collect clean signals. At the time, we weren’t sure if it was the electrodes, the lack of contact with the scalp, or simply because each of us had hair, creating impedance (despite our assumption being that the patient/football player would be bald, which is best simulated by putting electrodes on the forehead). During our third progress report, we explained the issues we were running into and addressed how we planned on overcoming them, which entailed more testing and a potential second prototype.

Progress Report #4

To determine what the issue was, a teammate and I performed testing once again with the gold cup electrodes, the new electrodes (known as Dryodes), and a third set of electrodes known as Neuroline electrodes. Each electrode was functioning correctly, which can be seen below since the signals were between 25 to 75 microvolts.

Data from OpenBCI from electrodes (20).png

Therefore, we determined that we needed to increase contact in the helmet between the user and the electrodes and find a way to transfer our data from the software onto MATLAB so we could record and compare different test results. I went with a teammate to purchase a smaller football helmet, which the electrodes team used to implement the Neuroline electrodes (seen in the image below).

Additionally, to conserve more space within the helmet, I researched and purchased two alternative battery solutions for the hardware since our current set up entailed using four AA batteries, taking up a bit of space. I also discovered how to transfer our EEG information and recording on the Cyton board’s software onto MATLAB, which enabled our lab mentor and PI to conduct post-processing analysis on our data.

The electrodes team also conducted phantom head testing with the different electrodes, which entailed testing each of the electrodes on a gelatin based phantom head, to verify how well the electrodes reacted to stimuli and picked up signals.

It turned out that the tighter helmet still did not provide sufficient compression on the scalp, so we brainstormed and considered potentially adding a pump to the helmet to increase compression and using a rugby cap within the helmet, which sits much tighter on the scalp (seen below).

Compilation of Semester

After presenting our final progress report, we were assigned with compiling all our information into a PowerPoint and a technical document to help future students understand where we left off so that they could continue the project. I helped compile our information for the PowerPoint and took responsibility in completing the technical document, detailing how to download the OpenBCI software, get the hardware to function, record brainwaves, save and export the information onto MATLAB, how to save data onto an SD card, and including potentially useful links and notes for future students. The PowerPoint presentation and technical document can be seen below. Note, the document was written in a more casual, conversational style to appeal more to future undergrad students who would be reading it.

EEG Spring 2022 Final Presentation

OpenBCI Technical Document

Results

Here is a summary of what we achieved as a group this semester

Established an electronic solution to viewing, capturing, and recording EEG signals
Tested three different electrodes and validated their ability to meet required microvolt range (25 to 75 microvolts) via OpenBCI software
Discovered how to import EEG data onto MATLAB for data processing
Created two prototype helmets using two different electrodes per helmet
Innovated ways to compress the electrodes onto the patient’s head for better connection and how to orient the electrodes within the helmet to capture appropriate signals
Phantom tested the electrodes to further verify their ability to pick up EEG signals
Established alternative battery solutions to conserve space within the helmets
Created a phantom head for future testing

Shortcomings & Lessons Learned

Towards the end of the project before we gave our final presentation, the group joked about how similar this experience was to a rollercoaster. Every time we accomplished something and reached a “high point”, it was almost immediately followed by a “low point”. For example, we would collect clean brainwaves using certain electrodes one day, but the next time we tested the same electrodes the brainwaves would be noisy. Another example was when we finally created our first prototype (and even our second prototype) only to realize that placing it on one of our teammates resulted in noisy brainwaves. By using the phantom head, we were able to prove that if the person putting on the helmet was bald, which was our original assumption, we would have been able to pick up clean brainwaves (in theory). However, since no one was willing to cut all their hair for the cause, it resulted in noisy brainwaves due to hair impedance. Nonetheless, it was discouraging to see our prototypes not function correctly after putting in the work to create them.

There were several times where we would progress only to find another problem come up, creating moments of confusion and frustration. However, we continued and persevered, which was a lesson this project taught me. The nature of every project is that it is made up of highs and lows. Therefore, when you hit a low, you need to continue and wait it out. Despite feeling discouragement when our results were not what we expected or wanted, we would try again, brainstorm alternate ways to progress, and waited for that next high point or accomplishment, which always came.

Another shortcoming we experienced was brainstorming unorthodox methods of approaching the problem too late into the project. In other words, my idea of cutting into the padding and installing the electrodes into those cut paddings was the only main idea we pursued. We brainstormed many ways on how to carry out that idea, but we limited ourselves to that idea for the majority of the project. It wasn’t until the end where one of our teammates brought up the idea of using a rugby cap inside the football helmet or using a pump with special padding to increase compression in the helmet. It would have been beneficial for us to have come up with some of these "out of the box” ideas early on and pursue them as alternate solutions, which could have inspired even more interesting ideas.

Overall Experience & Skills Attained

This project provided me with a very unique experience since it allowed me to work in a cross-functional team. The team was comprised of four biomedical engineers, one electrical engineer, and one mechanical engineer, which was me. Up until this point, I had never worked on a project with teammates that were not mechanical engineers (apart from my Co-Op), so it was great to see how my teammates approached the problem and brainstormed ideas based on their disciplines/backgrounds. I enjoyed working with people outside of my field, and I learned to appreciate and welcome different approaches.

During this project, I gained an introduction into the world of EEG signal collection and what tools and methods are typically used to pick up electrical signals from a person’s brain. I gained more experience on how to conduct research and how to learn since I had to understand how the Cyton PCB worked and how to get its software functioning. Additionally, I had to solve the issue of getting brainwave data from the OpenBCI software onto MATLAB for my lab mentor to use along with creating a detailed technical document explaining the electronics side of this project.