EEG for Human and Rat
|2020 Engineering Capstone Project|
|Team Name||Engineers Expecting Greatness|
|Duration||Fall 2020 - Spring 2021|
- 1 Problem Definition
- 2 Design Considerations
- 3 Project Learning
- 4 Final Design
- 5 Team Members
- 6 Additional Documentation
The main issue with electroencephalogram devices today is that they are rather expensive. With devices being around $800 plus additional fees for subscription services to use the software. These costs add up very quickly. Due to this, many high schools don’t have the funds to provide students with access to these kinds of devices. In addition to this, many high schools are beginning to attempt to get their students interested in the STEM fields. This has increased demand for cheaper, more viable devices and projects that can appeal to high school students. The secondary goal for this project is to create another EEG device for rats, that is to be used as a non-invasive testing method.
The first part of this project stems from the need for more affordable resources for high school student. The average cost for an EEG is upwards of $800, which is much too expensive to be utilizes in most high schools. This is where we come in, we are trying to create an EEG device that can be made for under $100 that is able to collect accurate data, and aide the learning of students. The second part of this project comes from there not being a way to read a rat's brain waves without being incredibly invasive to the rat as well as endangering its life. We are out to create an EEG that is small enough and accurate enough to read a rat's brain waves without being overly invasive.
For this project, our final deliverables will be slightly different based on the species it is meant for. For the human edition of the EEG, we are hoping for a fully functional, reliable, brain wave measuring device that is under $100. For the rat edition of this project, we are expecting to finish with a fully functioning EEG device, that is non-invasive, and can be used to measure the brain waves of rats.
This version of the EEG has a few requirements that need to be met. This will be a very lightweight device that is more portable than most. It will also be made from neoprene and be flexible enough to fit almost any head. This will be nice for high schoolers to use because it is very universal. It also needs to be under $100 to make in order for it to be affordable for high schools to have access to.
The rat version of this project has a few different requirements. The main one is for the device to be non-invasive. As of now, there really are not any EEG devices out there that work for a rat brain. Most procedures involve implanting the EEG itself into the rats brain, which requires surgery and is dangerous for the rat. For this device, we will be able to simply place it on top of the rats head, and it will read the waves.
For this EEG, we changed up the design from the previous years. They have all gone with a rigid 3D printed design, but we are trying to find something more flexible. For this we decided to use neoprene and make it more of a "wig-cap" design to ensure a stronger connection. The neoprene reduces the amount of noise to to lack of shifting around, and seems to be the best design. For the rat version of this, the device will be significantly smaller than the human one. We will instead use only four electrodes, and make the process as non-invasive as possible.
Throughout the process of trying to come up with a new design for this project, we had a few different prototypes in mind. The initial direction that we were thinking of going with was using hydrogel. This was a very short lived idea because we soon realized that hydrogel is only good for more of a one-time-use. After this, we created an actual prototype based on what we were given to work with.
Expensive EEG Device
This is the actual EEG device that we were provided with by Dr. Kumar. This device is very accurate and works quite well, this being because it is the very expensive version of what we are trying to make for cheap. This device gives us the opportunity to compare the data that we receive from our headset to the one that we know works. This will make our data much more accurate because it will let us know weather we are on the right path or not.
How EEG Devices Work: The EEG headset will have various electrodes that lay on the scalp in a non-invasive fashion. These electrodes pick up on any electrical activity going on in the brain, and it will convert those collected signals and store the data. This is most commonly demonstrated in the form of a moving graph (resembling a sine wave), and the spikes will go up and down based on the amount of electrical activity going on within the brain. EEG Devices are important because they can be used to help diagnose conditions such as seizures, epilepsy, head injuries, dizziness, headaches, brain tumors and sleeping problems. It can also be used to confirm brain death.
Wig Cap Design
This is the second design that we thought of, and it was eventually the design we went with. We ended up creating a fully functional prototype of this design. It is a scuba cap made out of neoprene that we had purchased, and then we sewed the electrodes into the cap. There are many pros to this design:
- Flexible: This design allows more comfort on the heads of the users.
- One size fits all: This design can be used on almost any head. The flexibility allows it to be diverse in size.
- Secure to head: Due to the neoprene material, this cap fits secure to the head of any wearer.
- Reduces noise: There are usually many outside sources of noise, but the cap has a tight fit that secures the electrodes to the scalp without room to move
- Comfortable: This design was made to be comfortable as well as functional.
There were a number of issues that we ran into while developing a functional prototype for the human edition of the EEG device. One of the issues that we run into is that the E-Motive device can be difficult to use on a regular basis. The felt tips on the device are held in very loosely and they fall out when using it. it also has a very hard time connecting to the head when in use. The signals can tend to be very noisy. Another issue that we ran into was the functionality of the second stacked circuit we developed. This is something that we were unable t get to be fully functional in time for expo, but the concept for the design has been validated.
Sources of Error
There are a few main sources from which the main errors come from with our EEG device. These errors can make the data acquired inaccurate and not usable, and it is very important that we catch these errors and stop them at the source.
This is a very important step when using the device. After the saline solution is applied, the headset needs to sit for approximately five minutes. Once the headset sits, the devise needs to be calibrated using a breathing technique we developed in order to calm the mind and represent accurate data. Once the calibration is done, the device is ready to have data be recorded.
While using the device, there are also a few physical sources of error that would arise. The first source was long or coarse hair, this would cause the nodes to not have a very secure connection to the scalp. This would make it so that the signals were not picked up and the readings would be off. Second up is an excess of movement while the cap is on. This would cause the nodes to pick up a lot more noise and it would muddle the data.
The team designed an affordable electroencephalogram (EEG) device to be used by high school students for better awareness of the breadth of fields of STEM. This neurobiology device demonstrates the responsibility of different parts of the brain. The main requirement is to build a cheap ($120) device that high school students can easily build themselves, since all industry EEG devices cost $900, which is not feasible for public schools. The device can be assembled by sewing gold plated electrodes into a neoprene scuba wig cap.
The circuit card consists of 8 channels that amplifies the brain signals to the millivolt range, and filters them to cover 1 Hz to 100 Hz range, using surface mount components that are soldered into 2 layers. The software has no licensing cost, and can reliably demonstrate differences in brain signals from separate parts of the brain with the ability to tune how much gain is given into each channel allowing the user to amplify each channel and getting more control over the device to show differences in signals coming from different lobes of the brain at the same time.
The testing shows that the low-cost device can illustrate differences in brain functions in each lobe of the brain when the subject is exposed to different stimuli. The signals shown are similar to the signals of the industry used Emotive device. Continuing the testing of the device and refining it to include other features were the remaining sections of the project completion. These features included developing a game that shows the object jumping whenever brain signals are detected. After finishing our testing sessions, we concluded that this device could act as a fully functional low-cost EEG that shows interest in neurobiology and STEM fields.
For the rat edition of this project, our goal was initially to create an EEG device that would be non-invasive to a rat. Typical rat EEG devices are implanted in the brain of the rat which can be incredibly dangerous to the animals wellbeing.
As the year progressed, the team and our sponsors decided that it would be best to spend our time fully developing the human edition, so we were not able to make much progress on the rat edition. We were able to fully develop the concept as well as the circuit.
Conceptually, the main difference between the human and rat systems is that the rat would only have four nodes in the small headset, one to represent each part of the rats brain. The frequency for the rat circuit would also be significantly different from the human one.
For the future, the next group that works on this will have a very good jumping off point to develop this non-invasive EEG device for rats!
When designing the circuit board, our main focus was to maximize the educational value for high-school students to attract more eyes to the STEM fields. The circuit design is very simple and symmetric, which makes it more usable for educational purposes.
The circuit brings good practice with filters by attenuating all frequencies outside 1 Hz and 120 Hz. Moreover, a 2-layer circuit board is used for a more fun soldering experience, and better compatibility which allows the project to be wireless in the future work. In addition, we made use of surface mount and through hole components to allow students to work with both.
The circuit was able to tell the difference between brain signals while blinking vs brain signals with eyes closed. This was visualized through the Spike Recorder software and the new GUI program.
The new circuit design achieved filtering desired frequencies, but the output voltage was 150x lower than the original circuit card. Because of this, it was very difficult to get accurate readings on the GUI. With the male pin stacked design, it was found to be difficult to take apart for troubleshooting.
The purpose of the software was to create a visual and intuitive understanding of the headsets readings, as well as provide a module to extend the functionality to other programs. The code was also designed to be intuitive to anyone willing to read it, as the application is meant for educational purposes.
For this reason Python was chosen, which is an intuitive programming language with many existing scientific tools. Python has a framework of the Qt user interface called PyQt, which can quickly create a window displaying useful content using an object-oriented approach.
It also has the PyQtGraph library which can display live charts and graphs in considerably fast times. For these reasons the code was developed with these tools.
Major: Mechanical Engineering
Major: Electrical Engineering
Major: Biological Engineering
Major: Biological Engineering
Major: Computer Science
Major: Computer Science
9/15 Meeting Minutes
9/18 Meeting Minutes
9/22 Meeting Minutes
9/24 Meeting Minutes
9/29 Meeting Minutes
10/2 Meeting Minutes
10/6 Meeting Minutes
10/8 Meeting Minutes
10/9 Meeting Minutes
10/12 Meeting Minutes
10/13 Meeting Minutes
10/15 Meeting Minutes
10/20 Meeting Minutes
10/23 Meeting Minutes
10/27 Meeting Minutes
10/29 Meeting Minutes
11/3 Meeting Minutes
11/5 Meeting Minutes
11/9 Meeting Minutes
11/10 Meeting Minutes
11/12 Meeting Minutes
11/16 Meeting Minutes
11/17 Meeting Minutes
12/1 Meeting Minutes
12/3 Meeting Minutes
12/8 Meeting Minutes
1/15 Meeting Minutes
1/19 Meeting Minutes
1/22 Meeting Minutes
1/26 Meeting Minutes
2/2 Meeting Minutes
2/9 Meeting Minutes
2/23 Meeting Minutes
3/2 Meeting Minutes
3/23 Meeting Minutes
3/25 Meeting Minutes