[Mne_analysis] problem with mne_setup_source_space : Source space creation failed

Lucille Lecoutre lu.lecoutre at gmail.com
Thu May 23 13:02:07 EDT 2013
Search archives:

2013/5/23 Lucille Lecoutre <lu.lecoutre at gmail.com>

> Hi Dan,
>
> I am not totally sure if this is a problem or not for the group analyses,
> as I this is the first time I do source reconstruction with MEG. The only
> thing I could think about is that it could introduce a statistical bias in
> the grand averaging procedure given the different number of vertices.
> Here is my procedure after having computed the inverse model : I make a
> grand average of my source estimates (using  mne_average_estimates), and
> then selecting visual and auditory labels on the fsaverage brain. All the
> subsequent analysis are done within these labels.
> I haven't tried to go through the whole procedure with this particular
> subject porcessed with ico -6 so far. I should probably check first that
> mne_average_estimates accepts to average source estimates (produced with
> mne_python) having a different number of vertices.
>
> Thank you very much for your help,
>
> Lucille
>
>
>
>
> 2013/5/23 dgw <dgwakeman at gmail.com>
>
>> Hi Lucille,
>>
>> While it might be the preferable option, I cannot see a reason why it
>> should be necessary unless you are comparing different groups (like one
>> group with several participants with this issue and another that doesn't
>> have any like this). Assuming you do the morphing and smoothing
>> appropriately for this participant and for the others (that is more
>> smoothing according to the source space decimation), I would not have a
>> problem with this as a reviewer (though perhaps other readers may disagree:
>> please speak up if you can see a problem). It shouldn't introduce any error
>> (assuming the rest of your analysis is appropriate i.e. appropriate
>> looseness type orientation constraint etc.).
>>
>> Although, I guess given the automated nature, it is likely fairly simple
>> to re-generate your source space, forward model and inverse solutions for
>> this data (depending on how complex your analysis is).
>>
>> HTH,
>> D
>>
>>
>>
>>
>> On Thu, May 23, 2013 at 9:54 AM, Lucille Lecoutre <lu.lecoutre at gmail.com>wrote:
>>
>>> Hi Dan,
>>>
>>> Thank you very much for your reply.
>>> I guess I will then have to redo the analyses for each subject with the
>>> -6 option if I want to include this one in the group analysis...
>>>
>>> Lucille
>>>
>>>
>>>
>>>
>>> 2013/5/23 dgw <dgwakeman at gmail.com>
>>>
>>>> Hi Lucille,
>>>>
>>>> This is due to the nature of this participant's source space.
>>>> Unfortunately, the only remedies are to use a different ico setting (with a
>>>> lower number of vertices, for example -6) or to use the -spacing option).
>>>>
>>>>
>>>> HTH,
>>>> Dan
>>>>
>>>>
>>>> On Tue, May 21, 2013 at 10:00 AM, Lucille Lecoutre <
>>>> lu.lecoutre at gmail.com> wrote:
>>>>
>>>>> Dear All,
>>>>>
>>>>> I have a problem with the creation of the source space
>>>>> (mne_setup_source_space)  with the -ico 5 option for one particular
>>>>> subject. Here is the error reported by MNE :
>>>>>
>>>>> --------------------------------------------------------------
>>>>> $ mne_setup_source_space --ico 5 --overwrite
>>>>>
>>>>> Setting up the source space with the following parameters:
>>>>>
>>>>> SUBJECTS_DIR = /neurospin/meg/meg_tmp/AVclock_Lucille_2013/mri/
>>>>> Subject      = tc_120199
>>>>> Surface      = white
>>>>> Icosahedron subdivision grade 5
>>>>>
>>>>> >>> 1. Creating the source space file
>>>>> /neurospin/meg/meg_tmp/AVclock_Lucille_2013/mri//tc_120199/bem/tc_120199-ico-5-src.fif...
>>>>>
>>>>> mne_make_source_space version 2.6 compiled at May  6 2013 04:18:39
>>>>>
>>>>> Loading
>>>>> /neurospin/meg/meg_tmp/AVclock_Lucille_2013/mri//tc_120199/surf/lh.white...
>>>>> Triangle file : created by LLECOUTR on Wed May  1 02:26:12 2013 nvert
>>>>> = 136783 ntri = 273562
>>>>>     Triangle and vertex normals and neighboring triangles...[done]
>>>>>     Vertex neighbors...[done]
>>>>>     Distances between neighboring vertices...[820686 distances done]
>>>>> Doing the icosahedral vertex picking...
>>>>> Loading geometry from
>>>>> /neurospin/meg/meg_tmp/AVclock_Lucille_2013/mri//tc_120199/surf/lh.sphere...
>>>>> Triangle file : created by LLECOUTR on Wed May  1 03:29:58 2013 nvert
>>>>> = 136783 ntri = 273562
>>>>>     Triangle and vertex normals and neighboring triangles...[done]
>>>>>     Vertex neighbors...[done]
>>>>>     Distances between neighboring vertices...[61440 distances done]
>>>>> Mapping lh tc_120199 -> ico (5) ...[done]
>>>>> Setting up the triangulation for the decimated surface...
>>>>> loaded lh.white 10242/136783 selected to source space (ico = 5)
>>>>>
>>>>> Loading
>>>>> /neurospin/meg/meg_tmp/AVclock_Lucille_2013/mri//tc_120199/surf/rh.white...
>>>>> Triangle file : created by LLECOUTR on Wed May  1 04:55:58 2013 nvert
>>>>> = 137309 ntri = 274614
>>>>>     Triangle and vertex normals and neighboring triangles...[done]
>>>>>     Vertex neighbors...[done]
>>>>>     Distances between neighboring vertices...[823842 distances done]
>>>>> Doing the icosahedral vertex picking...
>>>>> Loading geometry from
>>>>> /neurospin/meg/meg_tmp/AVclock_Lucille_2013/mri//tc_120199/surf/rh.sphere...
>>>>> Triangle file : created by LLECOUTR on Wed May  1 06:15:31 2013 nvert
>>>>> = 137309 ntri = 274614
>>>>>     Triangle and vertex normals and neighboring triangles...[done]
>>>>>     Vertex neighbors...[done]
>>>>>     Distances between neighboring vertices...[61440 distances done]
>>>>> Mapping rh tc_120199 -> ico (5) ...[done]
>>>>> Source space vertex moved from 57859 to 55128 because of double
>>>>> occupation.
>>>>> Source space vertex moved from 57859 to 52619 because of double
>>>>> occupation.
>>>>> Source space vertex moved from 55128 to 52611 because of double
>>>>> occupation.
>>>>> Source space vertex moved from 52619 to 57845 because of double
>>>>> occupation.
>>>>> Source space vertex moved from 57845 to 57835 because of double
>>>>> occupation.
>>>>> Source space vertex moved from 57835 to 57846 because of double
>>>>> occupation.
>>>>> Source space vertex moved from 57860 to 59133 because of double
>>>>> occupation.
>>>>> Source space vertex moved from 59133 to 60383 because of double
>>>>> occupation.
>>>>> Source space vertex moved from 60383 to 59154 because of double
>>>>> occupation.
>>>>> Source space vertex moved from 52619 to 53812 because of double
>>>>> occupation.
>>>>> vertex 57859 is in the map multiple times.
>>>>> Cannot continue
>>>>> Source space creation failed
>>>>> Temporary files removed.
>>>>> ---------------------------------------------
>>>>>
>>>>>
>>>>>
>>>>> I saw that this problem has been reported before (see this thread<https://mail.nmr.mgh.harvard.edu/pipermail//mne_analysis/2011-November/000837.html>from November 2011).
>>>>> I tried as suggested by Alex to use the -ico -6 option and it works
>>>>> perfectlly well.The problem with that solution is that I want to work with
>>>>> the FreeSurfer average brain, all the other subjects have been processed
>>>>> with the ico 5 option.
>>>>>
>>>>> So here are my questions :
>>>>> Does it make sense at all to work on an fsaverage brain if the number
>>>>> of the sources are not the same across subjects?
>>>>> It is said in the previous thread that this a known problem, and that
>>>>> it is caused by "some safety checks run by MNE". Is there any way to bypass
>>>>> these sanity checks?
>>>>>
>>>>>
>>>>> Many thanks,
>>>>>
>>>>> Lucille Lecoutre
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> 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.
>>>>>
>>>>>
>>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20130523/40b78415/attachment.html 


More information about the Mne_analysis mailing list