Source localization

✉️ Mailing List Archive (read-only)
1

External Email - Use Caution

Dear MNE community,
I came up with unusual results while reconstructing the sources using
standard forward model. The resulting topography of sources is in the
attachments (mean amplitude for one person). Forward model was computed
with 3-layer bem model with oct5 spacing for 2052 sources, for inverse
solution the 'fsaverage' trans-matrix was used. I also used standard
'easycap-M1' montage. The original EEG data was pre-processed via
BrainVision Analyzer and looks nice at sensor space topomaps. I have
already looked for errors related to localization, forward or inverse
computations, but still did not find any. Has anyone come across such
results before?
I believe it should be solvable.
Any ideas?
Thanks in advance!
Ann

2

External Email - Use Caution

Dear Ann,

can you share an image of a sensitivity map for your forward problem?

https://mne.tools/stable/auto_examples/forward/plot_forward_sensitivity_maps.html

did you look at the whitened data?

https://mne.tools/dev/auto_tutorials/evoked/plot_whitened.html#evoked-data-with-whitening

can you share a figure ?

Alex

3

External Email - Use Caution

Dear Ann,

this looks like ?spotty brain decease?, caused by visualizing the source data on the wrong brain model. Are you 100% sure that the model you use for visualization is the same model for which the source space was constructed?

best,
Marijn.