The new BrainVision Recording Format (BVRF) and how to use it

by Dr. Alex Kreilinger
Technical Product Manager (Brain Products)

With the release of BrainVision Recorder 2, we also updated our recording file format from the original BrainVision Core Data Format (BVCDF) to the new BrainVision Recording Format (BVRF). The new modern format provides new features such as meta data in JSON format or support for multiple participants recordings while keeping the output BIDS-friendly for easy file conversion. Here is how to access the newly recorded files and how to keep working with data in new and established ways.

The new BVRF makes sure you are prepared for the future

While the binary data format remains the same at its core, we added several changes that make it easier to work with the new file format, and to provide important new functionalities. These are some of the main changes:

  • Metadata files based on JSON. The JSON format is easily readable by human and machine and is therefore perfectly suited for storing important contextual information that needs to be preserved in parallel to continuous data signals.

  • An additional but optional impedance file allows the collection of several impedance values over time, better separated and easier to read compared to being saved in the header file in the original BVCDF.

  • We follow the guidelines of the BIDS standard, so data can be converted and shared among researchers as easily as possible.

  • Acknowledging the growing interest in hyperscanning experiments. To offer a suitable file format for such experiments, writing the signals of multiple participants directly in one single file is already supported. Data from different people can be analyzed simultaneously, while common signals (for example from sensors) can still be assigned to everyone at the same time.

BrainVision Recorder 2 already uses the new file format, but recording to the new format is only the first step. It is equally important to have the right tools for accessing the data conveniently and reliably. Therefore, the new BVRF format can be directly loaded in BrainVision Analyzer from version 2.3.1 which can be downloaded here.

Not using BrainVision Analyzer for your data analysis? – No worries!

Because we know that researchers are frequently using other tools outside of the Brain Products portfolio to analyze data recorded with Brain Products amplifiers, we also made sure to provide alternatives:

  • For our Python users, we are happy to inform you about a newly developed package called PyBVRF. It is available for download on GitHub and can easily be imported into your standard MNE routines. If you prefer to work with a standalone graphical user interface (GUI), there is also an option, as the new PyBVRF package is already integrated seamlessly in MNELAB from version 1.3.1. Using these tools, you can load and analyze data recorded in the new format just as you are used to with the original BVCDF with the added features as explained above.

  • For our MATLAB® users, we provide the BrainVision BVRF reader directly on our own GitHub page. Please read the documentation on GitHub for information on how to use it in EEGLAB or directly in MATLAB. In EEGLAB, you can also directly add the plugin from its own extensions manager. Once the extension is added, you can easily import BVRF files directly into EEGLAB via the GUI.

Compatibility with the BrainVision Core Data Format (BVCDF)

If you are concerned about mixing recordings from Recorder 1 and Recorder 2, you can easily convert the new format to BVCDF. In Analyzer, you can use the built-in export function (see the BVRF support article for detailed instructions). In the previously mentioned MATLAB and Python solutions, the currently implemented export functions still save to BVCDF, as Recorder 2 is currently the only software saving to BVRF directly.

Still hesitating about switching over to the new format? – Give it a try!

Why not give it a try and download some demo data, we recorded for you?

Download Demo Data

The data was recorded during a passive visual oddball paradigm with an actiCHamp Plus amplifier and 32 actiCAP slim electrodes at 1000 Hz sampling rate. In addition to EEG signals, we recorded the sensor signal from a photo sensor that is time-locked to visual cues, and you can see some recorded impedance data in the additional impedance files.

Take a look at the results of our basic analysis of generated ERPs and P300s, or try other things with the data and meta data provided. The demo set also includes a history template that you can apply to the raw data. Feel free to change the settings to your liking.

The new BrainVision Recording Format (BVRF) and how to use it

Raw data before filtering and re-referencing, including the photo signal and markers
for frequent (‘S 1’) and rare (‘S 2’) stimuli.

The new BrainVision Recording Format (BVRF) and how to use it

Averaged responses to visual stimuli.
Frequent events (blue), rare events (red), and difference (black).

The new BVRF is an important step into the future of EEG recordings, and we’re eager to hear what you think. We welcome you to share your thoughts and experiences with the new format with us at techsup@brainproducts.com.

Most recent articles