Add clinical data in a standardized way

1

Hi all,

are there guidelines to include clinical information with in BIDS? So far I name everything under participants. In my case, it is DBS data. This means that with e.g. electrode contact, stimulation frequency… very manifold information can be listed.

Greetings,
Matthias

2

I would start with iEEG and see if you can do most of what you want just by adding additional columns to the file types defined there: Intracranial Electroencephalography - Brain Imaging Data Structure v1.8.0

3

Thanks for the quick reply. Can the field ‘Electrical stimulation’ be added to other modalities like MEG or EEG? In my case, I’m using a MEG scanner, and the measurement is performed with DBS turned ON and OFF. Opening a field for iEEG just for the electrical stimulation seems a bit confusing to me. If the field ‘Electrical stimulation’ can’t be added to MEG, I think it is more convenient to add the DBS information to the participant’s files.

4

Generally, simultaneous iEEG/MEG is permitted. I don’t know the details of how people combine things, but I would expect metadata that is permissible for iEEG to be permissible for an MEG recording when there are iEEG electrodes. See Magnetoencephalography - Brain Imaging Data Structure v1.8.0

Note that if EEG is recorded with a separate amplifier, it SHOULD be stored separately under a new /eeg data type (see the EEG specification).

If however EEG is recorded simultaneously with the same MEG system, it MAY be stored under the /meg data type. In that case, it SHOULD have the same sampling frequency as MEG (see SamplingFrequency field below). Furthermore, the EEG sensor coordinates SHOULD be specified using MEG-specific coordinate systems (see coordinates section below and the Coordinate Systems Appendix).

I would expect the same conditions to apply to your case.

5

Hi,
thank you very much for the suggestion. However, in my case, the problem is that there are no iEEG electrodes. My dataset is a pure MEG dataset, but where the patient’s DBS system is switched on and off in between. Thus, I have not found it helpful to create an iEEG sidecar. In my case, I could also store the stimulation parameters in the participants.tsv file, but as soon as different stimulation parameters are used in the different runs, this would not really be practical, in my opinion. Would it then make more sense to create an iEEG sidecar?

6

I think you’ve hit the limits of my understanding of the electrophysiological context. I don’t have much in the way of advice here. I would generally say: Organize the data as makes sense to you within the context of BIDS. If you’re able to mix and match pieces without breaking validation, that’s good. You may want to consult with the makers of tools for processing the type of data you have about if there’s a convention they’ve adopted. If that maker is you and there’s nobody else processing the same type of data, your need for consultation is probably limited.

7

Thanks for the ideas, I will try to stick to the BIDS specifications as much as possible.

One thing that might help is if the electrical Stimulation field of the iEEG specifications Intracranial Electroencephalography - Brain Imaging Data Structure v1.8.0 could be added for MEG as well. Would something like that be possible?

8

Yes, I think that would make sense. The way to start that discussion is to open an issue on the specification repository.

1 Like