If you want to digitize brainwaves and electroencephalograph (EEG) data, while interfacing with other Arduino hardware and software, this is the Arduino shield for you!

HackEEG is an affordable, high-performance, fully open source TI ADS1299 Arduino shield ideally suited for digitizing biosignals like EEG, EMG, and EKG. If you want to read your brainwaves, do brain-computer interfacing, or build high-performance lab instruments for neuroscience or biology, HackEEG is for you! We’ve spent the last few years partnering with major research institutions and pharmaceutical companies in the US and Europe and now we want to bring HackEEG to you!

HackEEG shield - angled

This shield is a field-proven, high-performance, open-source lab instrument. All hardware, drivers, and Python client software are completely open source and easily modified. HackEEG is ideally suited for neuroscience biosignals applications – EEG, EKG, EMG, and more.

Features and Specifications

• Eight analog-digital conversion (ADC) channels, each with a 24x programmable-gain amplifier (PGA).
• Up to four shields can be stacked on one Arduino Due, for a total of 32 EEG channels
• DMA drivers can read and transfer eight channels of 24-bit resolution data at up to 16,000 samples per second (the maximum rate of the ADS1299 chip). 32 channels can be read at 4,000 samples per second.
• USB 2.0 HS (480 Mbps) connectivity when using Arduino Due
• Python client software and APIs provide easy interchange of real-time data to Lab Streaming Layer, pandas, numpy, and more.
• Data can be streamed via Lab Streaming Layer at 16,000 samples per second from a laptop.
• SPI EEPROM for storing configuration data
• Jumpers for configuring most of the data and control lines to the Arduino
• Jumpers for configuring on-board instrumentation power supplies to output -2.5 V to 2.5 V or 0 V to 5 V
• Also compatible with the Arduino Mega 2560 but with reduced sampling speeds. 5 V to 3.3 V voltage-level shifters included on board.
• Open Source Hardware certified

High-performance

HackEEG allows you to access the entire range of the TI ADS1299 EEG system-on-a-chip‘s features and performance. The Arduino drivers and Python client software allow complete access to the ADS1299 registers, and can read 8 channels of 24-bit ADC data at the maximum rate of the chip, 16,000 samples per second.

On-board instrumentation power supplies can be configured for 2.5 V to -2.5 V or 5 V to 0 V. Single-ended and differential modes are both possible with P and N inputs available on the HackEEG’s electrode connector.

HackEEG, subject, and technician

8, 16, 24, or 32 channels

Up to four boards can be stacked on one Arduino Due for 8, 16, 24, or 32 channels.

Hackable

Since this is an Arduino shield, you can stack it with existing shields, or add hardware of your own design. For example, we have users who added analog preamplifiers and custom image digitization synchronization hardware on their own custom shields.

All the software is completely open, so if you want to change it or add to it, you can! The HackEEG Arduino driver is text-mode, and has a JSON Lines protocol, so if you want to write client software in another language, you can. There is also a compact binary format that uses MessagePack that is used just for data streaming. Libraries for both JSON Lines and MessagePack data formats are available in many computer languages like C#, Javascript, C++, Ruby, Haskell, and others.

Lab Instrument

This is an actual lab instrument being used to do science in many research and pharmaceutical labs. It has high-resolution, high-bandwidth data acquisition capabilities with very low noise characteristics; even though it’s only $335.00, HackEEG matches the performance capabilities of closed-source amplifiers costing many thousands of dollars, and it can’t be beat for openness and programmability.

HackEEG has a wired interface so can be used directly connected to a laptop or desktop computer to do high-speed capture of biosignals. With the addition of a Raspberry Pi 4, it can send data wirelessly via Wi-Fi.

Lab Streaming Layer

HackEEG’s Python client software can stream data via the standard neuroscience software communication protocol Lab Streaming Layer either on a single laptop computer or via Wi-Fi. This means you can stream HackEEG directly to your favorite neuroscience software or Matlab running on your laptop or remote computers, record data streams, and integrate HackEEG data with other data streams.

HackEEG connected to a Raspberry Pi 4 sending Lab Streaming Layer data over WiFi

OpenBCI Compatible

Using our fork of OpenBCI, you can stream HackEEG to OpenBCI and visualize it there. (We hope to get our changes merged to mainline.)

OpenBCI Lab Streaming Layer datasource selection:

OpenBCI Lab Streaming Layer source selection

OpenBCI showing HackEEG data being streamed by Lab Streaming Layer:

OpenBCI showing HackEEG Lab Streaming Layer data

Field-proven

HackEEG is being successfully used for neuroscience research by these organizations:

Comparisons

Manufacturing Plan

Starcat has manufactured more than 100 of the current revision of HackEEG with our current manufacturer and it is being used successfully at major research institutions and pharmaceutical companies. Printed circuit boards (PCBs) are manufactured in China; assembly is done in the USA. If there are more than 300 orders, we may have the boards made by one of the manufacturer’s facilities in Mexico.

We’ve used this manufacturer for more than four years and they have been highly reliable.

Shipping and Fulfillment

• Orders will ship from Crowd Supply's warehouse in the US.
• Domestic orders ship for free. Due to the complexities and costs of international shipping, an international surcharge of $15 will be assessed at checkout.
• For more information, please read this Guide page about ordering, paying, and shipping.

Risks and Challenges

This is not a device made for consumer home use. The internal electronics are exposed during normal operation to facilitate easy access by developers and professionals. As such, you must be aware of the hazards of operating exposed electronics, and are familiar with standard anti-static procedures for handling bare electrical components or be prepared to make your own case.

There is no patient isolation circuitry on the device, so you will need to be sure to operate the HackEEG and all connected devices from battery power.

As with any manufactured product, there are risks inherent in the process of sourcing components and shipping finished product to customers. That said, we are very experienced in this process, have used the manufacturer before on several projects including beta runs of the identical board, and don’t anticipate running into any problems we won’t be able to overcome.

Support and Documentation

All hardware, drivers, and software are open source and available on Github:

• Autodesk Eagle Design files and hardware documentation are available on Github (Cern OHL 1.2)
• Connector pinouts
• Configuration settings reference, commonly-used settings
• Schematic (PDF)
• Driver files (Arduino sketch and Python client software) on Github – Driver is GPL3; Python client software will have an Apache License once the campaign completes.
• OpenBCI_GUI software that takes Lab Streaming Layer input - a fork of the great OpenBCI_GUI
• TI ADS1299 datasheet – the EEG system-on-a-chip used in this shield

Notes

In some of the photos, the HackEEG DIN 42-802 electrode connector board has nine connectors: eight inputs, one ground. The production connector board will have ten DIN 42-802 connectors: eight inputs, one reference, one ground. The videos were shot with a variant of the connector board that had the reference electrode connector placed on the back of the board.

Important Warning

Although HackEEG has some of the features of a medical device, it does NOT have any certifications (FDA, CE, IEC, etc.) and is NOT officially approved for medical or diagnostic use. It is your responsibility to ensure your safety when using the device. Furthermore, you should never power the device from a non-isolated power source.