[Mne_analysis] mne_setup_source_space

[Gustavo Sudre]

Alex,

Sure thing. I don't have the sphere-inflated versions of these surfaces, so I've been using the spacing option. For example:

mne_setup_source_space --subject mni --surface cortex --overwrite --spacing 3

is the most decimation I can get:

[…]
loaded lh.cortex 2562/40962 selected to source space (approx. spacing = 3 mm)
[…]

Any other spacing bigger than that gives me the same numbers. For example:

[…]
loaded lh.cortex 2562/40962 selected to source space (approx. spacing = 20 mm)
[…]

If I try a different surface, I can get to a smaller (extreme) number of sources, which made me wonder if there was anything specific in the decimation algorithm about the ratio of decimated / original vertices:

mne_setup_source_space --subject mni --surface thalamus --overwrite --spacing 3
[…]
loaded lh.thalamus 352/3016 selected to source space (approx. spacing = 3 mm)
[…]

but

[…]
loaded lh.thalamus 9/3016 selected to source space (approx. spacing = 20 mm)
[…]

and I can go all the way down to 3 sources… not that I'd do that.

Let me know if you need more details.

Best,

Gus

On Jun 5, 2013, at 11:32 AM, Alexandre Gramfort <gramfort at nmr.mgh.harvard.edu> wrote:

> hi Gus,
> 
> can you give the lines you typed and be more specific about ico or oct
> parameters you used.
> 
> A
> 
> On Wed, Jun 5, 2013 at 4:12 PM, Gustavo Sudre <gsudre at pobox.com> wrote:
>> MNE-ers,
>> 
>> I was wondering if you could clarify what's the cause of the following behavior. I have a triangular mesh surface (.surf) that was created using mne_convert_surfaces from a .tri file. Now, I'd like to only use some of the points on the surface (i.e. decimate it), so I'm using mne_setup_source_space for that. I get a reduction in the number of sources up to spacing = 3 (from 40962 sources to 2562), but anything higher than that gives me the same number of sources. If I plot the "poka-dot" surface, I can see that there is still plenty of room for further decimation.
>> 
>> Is there anything in the decimation algorithm that prevents further decimation after a certain ratio? Or should I be checking my surface for some specific malformations?
>> 
>> Note that this is not a MEG problem, so I'm using the terminology of sources mostly to make it easier. The actual issue is simply decimating a surface. I'm just using the MNE tools because I'm familiar with them and this way I don't have to re-code these procedures :) I thought about implementing my own decimation, but if you know of standard algorithms to decimate a surface, preferably already implemented in Python, I'd appreciate it if you could point me to it.
>> 
>> Best,
>> 
>> Gus
>> 
>> 
>> 
>> _______________________________________________
>> Mne_analysis mailing list
>> Mne_analysis at nmr.mgh.harvard.edu
>> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis
>> 
>> 
>> The information in this e-mail is intended only for the person to whom it is
>> addressed. If you believe this e-mail was sent to you in error and the e-mail
>> contains patient information, please contact the Partners Compliance HelpLine at
>> http://www.partners.org/complianceline . If the e-mail was sent to you in error
>> but does not contain patient information, please contact the sender and properly
>> dispose of the e-mail.
>>