Elata EEG
June 2025 status update: Elata EEG is in early prototyping. Currently able to read/write registers over SPI; next step is real-time data capture and signal processing.
Project overview
Our first hardware project is our open source EEG. We are targeting a retail price of below $500 for an 8 channel EEG, which will prove a competitive price in the current market. For comparison, projects tend to cost several thousand dollars, are not open source, and cannot be as easily maintained. We will offer our EEG first as a kit, then as a hobbyist project, and finally as a complete commercially available unit for researchers. Cash flows from sales of the kit, priced at a fair markup, will untimely support our viability as an organization.
The main computational unit for the EEG is a Raspberry Pi. Using a Raspberry Pi will make this project substantially easier for a distributed set of participants to join our project. This will also allow Elata contributors even without a physical device to participate, as they can use any other Mac or Linux device when testing. Raspberry Pi's are also easily purchased online and will allow for later customization and extensibility as additional hardware projects progress.
Apart from the Raspberry Pi, the additional hardware used is the Texas Instruments ADS1299-based amplifier board, which is used for converting the analog electrical signal into a fully digital signal that can be sent to the pins on the Raspberry Pi board. Software for signals processing, and other core software, is written with the Rust programming language. Although relatively new, the Rust language has earned a stellar reputation as a systems language for type and memory safety that will make the device more reliable and less prone to error. There is also a very strong open source community of Rust developers from which we plan on working with more. Apart from Rust, the GUI (Graphical User Interface) for the touch screen is written in TypeScript, as it is the most widely preferred language for writing GUIs. Other ancillary scripts needed to run the Raspberry Pi may be written in Python or Bash.
We are also planning an upcoming module system capable of running Python code so that AI/ML models can be run on data in the device with a GUI by non-technical users.
Our initial prototype is currently operational and capable of producing EEG scans and saving them. It operates as a kiosk stand with a 5 inch touchscreen. Progress thus far has been extensively documented on X, Telegram, Discord, and elsewhere within the Elata community. As of writing (June 2025), the unit has been placed in a 3d printed unit, and any participants can 3d print their own device. We are currently working on extensively testing the device, finalizing Computer Aided Design files for the 3d printing enclosure, preparing initial sale of the device as hobbyist kits, and preparing a variety of partners with which to conduct studies with the device.
While our initial EEG will be 8 channels, we are also working on business development initiatives to develop higher channel models - with the end goal of a 32 channel eeg, which is considered to be the 'gold standard' for research purposes - pending research initiatives on increasingly complex psychiatric conditions. While we believe that an 8 channel EEG is an ideal target to begin these initiatives, we believe that in the long run we will want our modular architecture to benefit from selling units with higher numbers of channels, as they can accommodate more robust research projects.
Technical details
Setup guide
git clone https://github.com/Elata-Biosciences/elata-eeg
cd elata-eeg
chmod +x install.sh
bash install.shDeveloper usage
# Stop kiosk mode
bash stop.sh
# Term 1, driver
cd driver; cargo build
# Term 2, daemon
cd daemon; cargo build; cargo run
# Term 3, kiosk
cd kiosk; npm run devRebuild production
# Stop
bash stop.sh
# ...<Change code here> ...
# Rebuild code base and run kiosk mode
bash rebuild.shLast updated