Combining EEG and MEG for source analysis

Dear all,

Just a brief question. The Brainstorm tutorials advocate against combining MEG and EEG for source analyses due to, e.g. sensitivity of the two modalities to different cortical regions and so on. However, the Brainstorm program itself is capable of doing this, but warn that the Brainstorm inverse models do not properly handle the combination of MEG and EEG yet.

Thus, were I interested in combining MEG and EEG, what sort of inaccuracies would I expect from running the analysis in Brainstorm? Would it be more appropriate to run the combined MEG and EEG source analyses in, for example, MNE python?

1 Like

There is nothing wrong with running the joint analysis in Brainstorm. The warning is just there to mention that the combination of both datasets is done in a simplistic manner in the software. You can also perform EEG and MEG source analysis separately in Brainstorm and then just add the respective results. Comparing with alternatives like MNE-python can be indeed interesting. We haven’t investigated this question too far actually. Let us know what you see!

Hi Brainstorm team,

I am in a hospital team performing simultaneous recordings with MEG(Elekta 306 channels) and HD-EEG(EGI 256 channels) on children with epilepsy. I am doing source localization with brainstorm for MEG and EEG, I manually detect interictal spikes and then do localization but I am only able to do them separately and it does not let me run them together. I want to prove that combined use of EEG with MEG improves source localization. Does Brainstorm have a feature like that? If you can provide me some insight I would be very grateful! Thank you in advance

We give recommendations regarding the combined EEG+MEG source estimation in our 2019 article about group analysis, but indeed this is missing from the online tutorials:

Magnetoencephalography and EEG sensor data can be processed jointly to produce combined source estimates. Joint processing presents unique challenges because EEG and MEG use head models that exhibit differing sensitivities to modeling errors, which can in turn lead to inconsistencies between EEG and MEG with respect to the (common) source model. In practice joint processing is relatively rare (Baillet et al., 1999). However, these data are complementary, which means that joint processing can potentially yield insights that cannot be seen with either modality alone. For example, in the evoked responses in the data set used here, the first peak over the occipital areas is observed in MEG (90 ms) slightly before EEG (110 ms). This delay is too large to be caused by acquisition imprecisions. This indicates that we are not capturing the same brain processes with the two modalities, possibly because the orientation and type of activity in the underlying cortical sources is different. MEG and EEG have different sensitivities to source orientation and depth. Given the challenges of joint processing, our advice is to first look at the source reconstructions for the two modalities separately before trying to use any type of fusion technique. In the following, since our goal is to illustrate an end-to-end processing pipeline rather than comprehensively demonstrate all of Brainstorm’s features, we restrict our inverse results to MEG-only processing and do not present results either for EEG alone or joint MEG/EEG processing.

but I am only able to do them separately and it does not let me run them together.

Note that Brainstorm should let you run the MEG+EEG computation if you want to. You should just get a warning:

I added a section about MEG+EEG source estimation in the tutorials:

@Sylvain @John_Mosher Do you have anything to add?