xdf_to_snirf

Translation of XDF files recorded with ARTINIS Starstim8 to SNIRF format.

The xdf format is a general format for storing time series data (https://github.com/sccn/xdf).
The snirf format is a format for storing nirs data (https://github.com/fNIRS/snirf).

Usage (without Git)

1. Download as a zip file (button < > Code)
2. Extract the archive on your computer (e.g., in your Download folder).
3. Move the extracted xdf_to_snirf folder to its intended location (e.g., in your Documents/CodeProjects/ directory)
4. In VSCode:
   1. Open the xdf_to_snirf folder in a new window.
   2. Open main.ipynb and click Run all
      • Running main.ipynb (without modifications) will convert the *.xdf files in the data/reference folder to *.snirf files in the results folder (it will create results).

Directory structure

The logic is to organize the data, code, and results in separate folders, all within the xdf_to_snirf folder. The structure is as follows:

xdf_to_snirf
   ├── readme.md     # description of the project 
   ├── data          # data used as input in the project  
   ├── results       # results of the conversion 
   ├── notebooks     # analyses with jupyter notebooks 
   ├── src           # python source code (alternative to notebooks)
   └── main.ipynb    # entry point: to run all analyses

Prerequisites

You need to have a Jupyter environment installed on your computer. For a minimal installation, refer to https://github.com/DenisMot/Python-for-HMS-Template.

Convert multiple XDF files to SNIRF format in batch

This is the main purpose of this project. To convert your own .xdf files, you need to set your own xdf_files list, which is the list of xdf files to convert.

Simple and easy way

The easiest way is to modify the call to get_xdf_files_in_directory() in main.ipynb. For example, if your .xdf files are located in /Users/denismot/my_data, you should modify it as follows:

xdf_files = get_xdf_files_in_directory('/Users/denismot/my_data')

Advanced way

If you want to convert .xdf files from several directories, you can set the xdf_files list manually. For example, if you want to convert the first two files in /Users/denismot/my_data, you can do as follows:

xdf_files = get_xdf_files_in_directory('/Users/denismot/my_data')
xdf_files = xdf_files[:2]

If you want to convert files from different directories, you can do as follows:

xdf_files = get_xdf_files_in_directory('/Users/denismot/my_data')
xdf_files += get_xdf_files_in_directory('/Users/denismot/my_other_data')

NOTE: The converted files will always be saved in the results folder. If the names of the xdf files from different directories are the same, the converted files will overwrite each other.

How it works

The conversion is done with the xdf_to_snirf function from the xdf_to_snirf module.
The xdf_to_snirf module is built from the latest version of the xdf2snirf notebook.
The xdf2snirf notebook includes tests (on reference data) that are not in the module.
The xdf2snirf notebook is a good place to start if you want to understand the conversion process.

Limitations

• The conversion is limited to the 8 NIRS channels of the ARTINIS Starstim8 device.
• The conversion includes the montage used in the NeuArm and ReARm projects (https://doi.org/10.1016/j.bbr.2023.114322).