Setting the EEG Montage¶
There are a number of pipeline operations that requires information about the sensor locations on the scalp (e.g. sensor locations are required for identifying neighboring sensors). Under the hood, the pipeline uses MNE-Python to set the EEG montage.
There are 3 separate fields that can be used for controlling montage setting
in the pipeline. all 3 of these fields exist under the project field of the
configuration file/class.
If your data are not already BIDS compliant, and you plan to use the pipeline
to bidfify your data, then you will want to set the bids_montage field. This will
set the montage of your EEG data as part of the BIDSification process, and will write
the data to an electrodes.tsv file, along with the BIDSified EEG data. In this case,
later steps in the pipeline will just use the aforementioned electrodes.tsv file for
retrieving information about sensor locations, and you don’t have to set a value in the
analysis_montage field.
If your data are already BIDS compliant, you can also leave analysis_montage field
blank, assuming there is an electrodes.tsv file for the pipeline to use.
If you wish to override this and use a specific montage, you are also allowed to
use one of the available montages in MNE-Python. You can find a list of options by
inspecting the output of mne.channels.get_builtin_montages().
If your EEG data has a custom montage that cannot be used with one of the outputs
of get_builtin_montages, you should set your montage and BIDSify your data before
running the pipeline.
Finally, the set_montage_kwargs field will accept any parameter that is accepted
by MNE-Python’s mne.Info.set_montage().
Example¶
project:
readme: "# Description of the dataset"
# Montage use to make file BIDS compliant.
# Can be path to digitized montage OR a string of one of mne's built in
# standard montages as specified by mne.channels.get_builtin_montages().
# Can be left empty if the input dataset is already in BIDS format.
bids_montage: GSN-HydroCel-129
# montage used while running the lossless pipeline.
# if empty, the pipeline will use the electrodes.tsv sidecar file, if created
# during the BIDS conversion.
# If specified, needs to be a string of one of mne's built in standard montages.
analysis_montage: ""
set_montage_kwargs: {}