TL;DR
Cerelog published an open-source 8-channel biosensing board (ESP-EEG) built around the ADS1299 ADC and an ESP32 module, aimed at EEG/EMG/ECG and BCI research. The project includes firmware, schematics, a forked OpenBCI GUI with Lab Streaming Layer support, and a Python client for streaming to neuroscience tools and Brainflow.
What happened
Cerelog has published the ESP-EEG, an open-source evaluation board that targets EEG, EMG, ECG and brain–computer interface research. The hardware pairs a Texas Instruments ADS1299 24-bit ADC for eight differential channels with an ESP32-WROOM-DA processor; default sampling is 250 samples per second. The repository provides firmware, schematics and a custom fork of the OpenBCI GUI that interoperates via the Lab Streaming Layer (LSL) protocol. A Python client and example scripts are included to stream data to other neuroscience software, MATLAB or the Brainflow API; a test plotting script is referenced. The design implements a closed-loop active bias (Drive Right Leg) approach intended to reduce common-mode noise and movement artifacts. The board supports USB-C and has Wi‑Fi hardware ready (Wi‑Fi support listed as beta). The project carries a critical safety warning: the device is not galvanically isolated and must be used only with battery-powered hosts; it is also not certified as a medical device.
Why it matters
- Provides an open, research-grade hardware and software stack that researchers and hobbyists can inspect, modify and integrate.
- Compatibility with OpenBCI GUI forks, LSL and Brainflow simplifies integration with common neuroscience workflows and analysis tools.
- Closed-loop active bias implementation aims to lower noise and reduce movement artifacts compared with passive bias approaches.
- Non-isolated design and lack of regulatory certification highlight safety and deployment limits for human use outside research settings.
Key facts
- Device: Cerelog ESP-EEG — an open-source 8-channel biosensing board for EEG/EMG/ECG/BCI research.
- ADC: Texas Instruments ADS1299, 24-bit, configured for eight differential channels plus one active bias channel.
- Processor: ESP32-WROOM-DA (dual-core, Wi‑Fi/Bluetooth capable).
- Default sample rate: 250 samples per second.
- Connectivity: USB-C for data/power; Wi‑Fi and Bluetooth hardware ready (Wi‑Fi support listed as beta).
- Noise reduction: Implements a true closed-loop active bias (Drive Right Leg) to cancel common-mode signals.
- Open-source: Firmware, software, schematics and example scripts are published on the project GitHub repository.
- Streaming: Python client uses Lab Streaming Layer (LSL) and can stream to neuroscience tools, MATLAB and Brainflow (custom instance available).
- Safety: Device is non-isolated — required battery operation; users must avoid connecting to mains-powered computers.
- Regulatory status: Not an FDA-cleared medical device and not UL or FCC certified; intended for engineering and research use only.
What to watch next
- Progress and stability of Wi‑Fi support (not confirmed in the source beyond 'beta testing').
- Whether the project pursues any regulatory certifications or hardware isolation revisions (not confirmed in the source).
- Potential upstream integration of the forked OpenBCI GUI or Brainflow custom instance into their mainline projects (not confirmed in the source).
Quick glossary
- ADS1299: A 24-bit analog-to-digital converter from Texas Instruments commonly used for high-resolution EEG and biopotential measurements.
- Lab Streaming Layer (LSL): A protocol and library for time-synchronized streaming of time-series data (e.g., biosignals) between applications and devices.
- Closed-loop active bias (Drive Right Leg): A technique that measures common-mode voltage on biopotential inputs, inverts it and actively drives that signal back into the body to reduce noise.
- Brain–computer interface (BCI): Technology that records and interprets neural signals to enable direct communication or control between the brain and external devices.
Reader FAQ
Is the ESP-EEG hardware and software open source?
Yes. The repository contains firmware, software, schematics and example scripts.
Can I connect the device to a wall-powered desktop for data collection?
No. The device is non-isolated and the project documentation warns that it must be used only with battery-powered hosts to avoid electric shock and ground-loop noise.
Is this a medical device or FDA-certified?
No. The product is intended for research, engineering and educational purposes and is not an FDA-cleared medical device; it is also not UL or FCC certified.
What software protocols are supported for streaming data?
The project provides a Python client and a forked OpenBCI GUI that use the Lab Streaming Layer (LSL); instructions for streaming to Brainflow are also provided.
What is the device's sample rate and channel count?
Default sample rate is 250 SPS and the board exposes eight differential channels plus one active bias channel.
Cerelog ESP-EEG High-precision 8-channel biosensing board designed for EEG, EMG, ECG, and Brain Computer Interface (BCI) research applications www.cerelog.com and Product Page: Here Need Help or have a question? Discord…
Sources
- Show HN: Open-Source 8-Ch BCI Board (ESP32 and ADS1299 and OpenBCI GUI)
- Show HN: Open-Source 8-Ch BCI Board (ESP32 and …
- This Open Source EEG Board Brings Real Brain-Computer …
- Cerelog ESP-EEG: 8-Channel Brain Computer Interface …
Related posts
- Palo Alto Linked to Possible $400M Bid for Israeli Startup Koi — Unconfirmed
- Qualcomm pushes into PCs with Snapdragon X2 Plus chips targeting mainstream
- Pioneer’s Sphera brings Dolby Atmos Spatial Audio to older cars via CarPlay