[FieldTrip] localspheres

Lilla.Magyari at mpi.nl Lilla.Magyari at mpi.nl
Wed Feb 6 10:43:38 CET 2013


hi Johanna and Allison,

I think there was a small bug in ft_prepare_headmodel
http://bugzilla.fcdonders.nl/show_bug.cgi?id=1679
a few weeks ago which caused this error. It has been fixed already, so if
the newest version of FT is used, ft_prepare_headmodel should run now with
cfg.method 'localspheres' without cfg.tissue specified.

But this depends on the tissues in the segmentation as Johanna wrote. But
besides 'brain' and 'scalp', it is also fine if the segmentation contains
the probability tissue maps of  the cerebro-spinal fluid, the gray and
white matter. In this case, brain binary mask will be created by the
function from csf,w hite and gray and the geometry of the brain tissue
will be used for the head-model.

I hope this helps too.

Best,
Lilla


> Sorry, I meant if you have a 'bnd'.   Of course you won't have the 'vol',
> that is the purpose of calling ft_prepare_headmodel!
>
> Johanna
>
>
> 2013/2/6 Johanna Zumer <johanna.zumer at gmail.com>
>
>> Dear Allison,
>>
>> Thank you for your question.  There have been some changes to
>> ft_prepare_headmodel recently (partly implemented by me) and my
>> apologies
>> that the documentation for these changes was not made thoroughly.   I
>> assume that you are refering to a very recent FT version.   :-)
>>
>> The way that you call ft_prepare_headmodel looks almost correct, except
>> that cfg.vol is not needed.
>>
>> If you have already a 'vol' or 'bnd' , then that can be used as the
>> second
>> input argument to ft_prepare_headmodel, and the 'segmentedmri' is not
>> needed.
>>
>> If you don't already have a 'vol' or 'bnd', then that will be created on
>> the fly within ft_prepare_headmodel from the segmentedmri that you
>> include
>> as second input.  However, what are the fields within 'segmentedmri'
>> that
>> you use?   If 'brain' or 'scalp' already exist, then those will be found
>> and used by default, and therefore a cfg.tissue is not needed.  However,
>> if
>> you have other fields with different names, etc then cfg.tissue should
>> point to the name of the subfield of segmentedmri that you wish to use
>> as
>> the headshape.   Please see ft_volumesegment for more details on which
>> tissue types or segmentation types can be computed as output and which
>> can
>> then be used in subsequent functions.   Perhaps simply recompute the
>> segmentedmri with the newest version of ft_volumesegment and then it
>> will
>> work?
>>
>> In the meantime, I will update the documentation to make all this more
>> clear.
>>
>> However, one more question.  I'm not sure what you mean by the error you
>> get with ft_getopt(cfg,'tissue').   If 'tissue' is not in cfg, then it
>> will
>> return a default empty variable, and there should not be an error.
>> Can
>> you say exactly what line of what function gives the error, and what the
>> error exactly is?
>>
>> And finally, when you successfully ran singleshell method, was that with
>> the segmentedmri (which would surprise me), or rather with a
>> cfg.headshape
>> option?
>>
>> Cheers,
>> Johanna
>>
>>
>>
>>
>>
>>
>>
>>
>> 2013/2/5 Nugent, Allison C. (NIH/NIMH) [E] <nugenta at mail.nih.gov>
>>
>>>  Hello, ****
>>>
>>> ** **
>>>
>>> I have what is possibly a naïve question, but I can’t seem to find
>>> anything on the tutorial pages.****
>>>
>>> ** **
>>>
>>> I’m trying to create a headmodel using the localspheres method, using
>>> the
>>> output of a segmented MRI.****
>>>
>>> ** **
>>>
>>> There is a tutorial page for using different types of headmodels, but
>>> it
>>> references ft_prepare_localspheres – which works, but it appears to
>>> have
>>> been deprecated.  When I try this:****
>>>
>>> ** **
>>>
>>> hdr=ft_read_header(‘mydata.ds’);****
>>>
>>> cfg=[];****
>>>
>>> cfg.method=’localspheres’;****
>>>
>>> cfg.grad=hdr.grad;****
>>>
>>> cfg.vol=segmented_mri;****
>>>
>>> vol=ft_prepare_headmodel(cfg, segmentedmri);****
>>>
>>> ** **
>>>
>>> I get an error at ft_getopt(cfg,’tissue’);****
>>>
>>> ** **
>>>
>>> Although the documentation for ft_prepare_headmodel states that the
>>> only
>>> additional field needed by localspheres is “cfg.grad” it seems that
>>> cfg.tissue may also be required?  If so, how do I set this – I can’t
>>> find
>>> any documentation that explains.****
>>>
>>> ** **
>>>
>>> I should note that I’ve run ft_prepare_headmodel successfully using a
>>> singleshell method.****
>>>
>>> ** **
>>>
>>> Thanks,****
>>>
>>> ** **
>>>
>>> Allison****
>>>
>>> ** **
>>>
>>> Allison Nugent, PhD****
>>>
>>> MRI Physicist****
>>>
>>> Experimental Therapeutics and Pathophysiology Branch****
>>>
>>> NIMH/NIH/DHHS****
>>>
>>> Ph. 301-451-8863****
>>>
>>> nugenta at mail.nih.gov****
>>>
>>> ** **
>>>
>>> _______________________________________________
>>> fieldtrip mailing list
>>> fieldtrip at donders.ru.nl
>>> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>>>
>>
>>
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip





More information about the fieldtrip mailing list