[Mne_analysis] Question about MNE source space patches

[Eric Larson]

        External Email - Use Caution        

ico6 is quite dense -- 40962 vertices per hemisphere. ico5 is already 10242
vertices per hemisphere, and oct6 -- which is the default in MNE-Python and
typically sufficient -- uses 4098 per hemisphere. Do you really need such a
dense source space? If so, you can probably get away without the `--cps`
option, since you'll have so many vertices the error of each normal
relative to the patch it represents will be diminished anyway. Then you
won't run into errors with assigning neighbors.

Eric




On Mon, Aug 24, 2020 at 3:09 PM Kadwani, Manorama <MKADWANI at mgh.harvard.edu>
wrote:

> Hi,
>
> I created the source space for a subject's BEM model using:
> mne_setup_source_space --subject <subject name> --ico -6 --cps --overwrite
>
> It completed with a message saying that "Defining patches (this will take
> a few mins)...Node 187 does not belong to any of the patches".
> It did not create the file name "oct-6p-src.fif" as it did for other
> subjects.
> I read that it's a file with patches information. What are patches and how
> to deal with the said node that doesn't belong to a patch?
>
> Thanks,
> Manorama
>
> _______________________________________________
> Mne_analysis mailing list
> Mne_analysis at nmr.mgh.harvard.edu
> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20200824/6527cd46/attachment.html