[Mne_analysis] equal length epochs & connectivity across epochs

[Phillip Alday]

The only EEG software that I'm aware of that will deal with unequal
length segments is FieldTrip. For most analyses, there is an implicit
assumption of equal length for both DSP and statistical reasons.

As far as implementing connectivity over time, you could in theory trick
MNE (or more precisely mne-connectivity as Alex indicated) to do this,
but appropriately transposing the data array in an Epochs object and
using the current method. I haven't thought about this means from an
inferential perspective, so I'm unsure how wise it would be to actually
do this. Adding this functionality to MNE could also take advantage of
the transpose-then-use-current-method, but somebody would have to take
the time to do this. (It's not going to be me.)

For the ragged segments / Epochs, it seems relatively unlikely to me
that anybody will implement this anytime soon because it involves
fundamentally changing the structure of the Epochs object, which
internally uses a Numpy (and hence rectangular, i.e. non ragged) array.

Hope this helps,

Phillip

On 11/25/21 03:59, chantal.oderbolz at uzh.ch wrote:
>
>         External Email - Use Caution        
>
> Dear MNE Python community,
>  
> I am currently using MNE Python to preprocess and calculate
> speech-to-brain connectivity. 
> I have come across the fact that in MNE epochs of equal length are
> required for further analyses (although I circumvented this problem
> when initially segmenting by using a for loop and saving individual
> epochs in a list) and that connectivity measures are implemented
> across trials and not over time within a trial. Both of these are the
> exact opposite of what I would like to do. 
> I have seen both issues raised on Github once or twice but was
> wondering whether this is something that is still on the table to
> implement in the future. I would also be interested in knowing whether
> anyone has dealt with this and, if so, how - maybe by customizing
> functions or similar? Or if I would need to migrate to a different
> software completely. 
> I look forward to your ideas and suggestions! 
>  
> Many thanks and best wishes,
> Chantal Oderbolz
>  
>
> _______________________________________________
> Mne_analysis mailing list
> Mne_analysis at nmr.mgh.harvard.edu
> https://secure-web.cisco.com/1Rz4-r3bEXahHXs7c9JAxGHYgkYsoNF4VNsQ0E28MyRskIP8IvRifVrdpI6gfq8kkYV4THyzZ4AmqDcWdb_ZCySGwjdyfJSD44Zebl-B62cgaybt9K3xFfr4A1fXaAPq7jm4Z0Fes4TJhM5QXf3n70Yy6NZ_3PsrmFyRwV007yYJSXm5hwusUysOFV4WpX_rkFv1yJgULQOKjCD53ByvtNf68XzxoVzzg7esZX7Aj-wfa2fmH8I4ZvWylOdpFGZss00u9RIVfYJpdMQgbd_O0C94rMCRCV-veOYBPQGwHRcN_jK0lJ8vDIWkCvxHZj8SOJXlH4DOfbvZi0SqlJJhUDw/https%3A%2F%2Fmail.nmr.mgh.harvard.edu%2Fmailman%2Flistinfo%2Fmne_analysis
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20211203/c0e1da7d/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_signature
Type: application/pgp-signature
Size: 495 bytes
Desc: OpenPGP digital signature
Url : http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20211203/c0e1da7d/attachment.bin 
-------------- next part --------------
        External Email - Use Caution        
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20211203/c0e1da7d/attachment-0001.html