[FieldTrip] Fwd: Coordsys issue ft_sourceplot

Mon Jun 20 21:22:10 CEST 2016

Dear Jan-Mathijs,

sorry that I bring up that post again, but I am still having problems with
the different coordinate systems in my data.

After replacing the .pos field in the source files with the one from the
"standard_sourcemodel3d5mm" ft_sourceplot works fine, i.e. the functional
data is not rotated anymore. However, the option cfg.atlas doesn't work
since I always get the error message "mismatch between the coord sys in the
atlas and the coord sys in the data...".

I checked the atlas, the functional data (stat) and the anatomy I use for
interpolation with ft_determine_coordsys. All three files have RAS
orientation. If I change the anatomical label of the axes of the functional
data, ft says "neuromag". Unfortunately, the function atlas_lookup works
only with "mni", "tal" or "spm". So I see where the error messages comes
from... The difference between MNI and neuromas is the unit (mm resp. m),
so I thought to use ft_convert_units, but that doesn't work for functional
data.

Is there any way that I can change the functional data to "MNI" (I thought
I did this already)? Sorry if this is a straight forward thing, but I don't
see where I make the mistake...

Thanks a lot and best wishes,

Andreas

2016-06-15 15:06 GMT+02:00 Schoffelen, J.M. (Jan Mathijs) <
jan.schoffelen at donders.ru.nl>:

> Are you sure that you use the pos-field from the template, i.e. from the
> standard_sourcemodel3d5mm?
>
> Best,
> Jan-Mathijs
>
> On 15 Jun 2016, at 14:31, Andreas Sauer <sauer.mpih at googlemail.com> wrote:
>
> Dear Jan,
>
> thank you for your email and the example code!
>
> I tried both ways to convert the coord systems, but unfortunately it still
> doesn't work.
>
> Since I used the tutorial procedure to create my sourcemodel (
> http://www.fieldtriptoolbox.org/tutorial/sourcemodel#subject-specific_grids_that_are_equivalent_across_subjects_in_normalized_space)
> I tried first with the easy way of only replacing the .pos field. But the
> plots don't change, the sources are still rotated by 90 degrees. Actually,
> the .pos fields are already the same before replacing it (see pic). Is
> there some error in my code?
>
>
> *% Step1: Load mri file*
> *mri =
> ft_read_mri('/data/home1/asauer/Projects/RPScZ/Analysis/Beamforming/MRI/HSM12_V2.mri');*
>
> *% Step 2: Segment the template brain and construct a volume conduction*
> *% model (i.e. head model): this is needed for the inside/outside
> detection of voxels*
> *cfg = [];*
> *cfg.output = 'brain';*
> *segmentedmri = ft_volumesegment(cfg,anatomy);*
>
> *cfg = [];*
> *cfg.method = 'singleshell';*
> *vol = ft_prepare_headmodel(cfg,segmentedmri);*
> *vol = ft_convert_units(vol,'m');*
> *figure; ft_plot_sens(sens,'style','*b'); hold on; ft_plot_vol(vol)*
>
> *% Step 3: Load sourcemodel *
> *load(fullfile(ft_path,'template/sourcemodel/standard_sourcemodel3d5mm'));*
> *template_grid = sourcemodel;*
> *clear sourcemodel*
>
> *% Step 4: Make sourcemodel*
> *cfg = [];*
> *cfg.grid.warpmni = 'yes';*
> *cfg.grid.template = template_grid;*
> *cfg.grid.nonlinear = 'yes';*
> *cfg.mri = mri;*
> *cfg.grid.unit = 'm';*
> *grid = ft_prepare_sourcemodel(cfg);*
> *figure; hold on;*
> *ft_plot_vol(vol,'facecolor','cortex','edgecolor','none'); alpha 0.5;
> camlight;*
> *ft_plot_mesh(grid.pos(grid.inside,:));*
>
>
> The only thing that is different from the tutorial is the unit
> conversion...
>
> After that I tried the transformation. All worked well until the source
> interpolation. ft_sourceinterpolation throws this error:
>
> *Attempted to access sel(1); index out of bounds because numel(sel)=0.*
>
> *Error in ft_sourceinterpolate>my_interpn (line 508)*
> *  ft_progress(sel(1)/num, 'interpolating %.1f%%\n', 100*sel(1)/num);*
>
> *Error in ft_sourceinterpolate (line 393)*
> *    interp.inside( sel) = my_interpn(double(functional.inside), ax(sel),
> ay(sel), az(sel), 'nearest', cfg.feedback);*
>
>
> which, to be honest, I don't really understand. Sorry...
>
> Best wishes,
>
> Andreas
>
>
> 2016-06-14 9:41 GMT+02:00 Schoffelen, J.M. (Jan Mathijs) <
> jan.schoffelen at donders.ru.nl>:
>
>> Dear Andreas,
>>
>> after having calculated the sources of a MMN paradigm (4D-system), I’d
>> like to plot the results on one of the anatomy template files in the
>> fieldtrip directory (single_subj_T1_1mm.nii) or alternatively on a MNI T1
>> template file (mni_icbm152_t1_tal_nlin_sym_09a.nii). Unfortunately, this is
>> not working well… As you can see on the attached picture, the functional
>> data is rotated by 90 degrees.
>>
>> I guess that something gets mixed because of the different coordsystems
>> of functional and anatomical data. The functional data is 4D, which means
>> ALS orientation. The anatomical is MNI, so RAS orientation. The plotting
>> works fine, if I use a mri file which was converted in field trip (V2.mri)
>> and therefore is also in ALS orientation. However, the anatomy doesn’t look
>> very pretty…
>>
>>
>> Is there any way that I can transform the functional data to RAS
>> orientation or the anatomical data to ALS orientation, respectively?
>>
>>
>> I tried with ft_volumerealign. However, since the template file has no
>> fiducials, it is hard to really precisely do this.
>>
>>
>> Fair point. I would however do it the other way around, using the V2 to
>> get the transformation from voxel space to MNI-RAS, and then use the
>> inverse of the transformation on the functional data: in this case you
>> could use ft_volumenormalise, with cfg.nonlinear = ‘no’. Alternatively, if
>> you are using this procedure:
>> http://www.fieldtriptoolbox.org/tutorial/sourcemodel#subject-specific_grids_that_are_equivalent_across_subjects_in_normalized_space,
>> you could directly replace the source.pos of your functional data with the
>> template.pos and then do the interpolation.
>>
>> To summarize:
>>
>> either:
>>
>> mri = ft_read_mri(‘V2.mri’);
>> mri.coordsys = ‘bti’;
>>
>> cfg = [];
>> cfg.nonlinear=’no’;
>> mrin = ft_volumenormalise(cfg, mri);
>>
>> transform_vox2bti = mri.transform;
>> transform_vox2spm = mrin.transform;
>> transform_bti2spm = transform_vox2spm/transform_vox2bti
>> source = ft_transform_geometry(transform_bti2spm, source);
>>
>> source_int = ft_sourceinterpolate(somecfg, source, niceanatomical);
>>
>> or:
>>
>> source.pos = templat.pos;
>> source_int = ft_sourceinterpolate(somecfg, source, niceanatomical);
>>
>> Best,
>> JM
>>
>>
>>
>> Any input on how to plot the sources on one of these nice anatomical
>> files is highly appreciated!
>>
>> Thanks and best regards,
>>
>> Andreas
>>
>>
>> --
>> Dipl.-Psych. Andreas Sauer
>> Max Planck Institute for Brain Research
>> Deutschordenstraße 46
>> 60528 Frankfurt am Main
>> Germany
>>
>> T: +49 69 96769 278
>> F: +49 69 96769 327
>> Email: sauer.mpih at gmail.com
>> www.brain.mpg.de
>> <Rotated_Source.png>_______________________________________________
>> fieldtrip mailing list
>> fieldtrip at donders.ru.nl
>> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>>
>>
>>
>> _______________________________________________
>> fieldtrip mailing list
>> fieldtrip at donders.ru.nl
>> https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>>
>
>
>
> --
> Dipl.-Psych. Andreas Sauer
> Max Planck Institute for Brain Research
> Deutschordenstraße 46
> 60528 Frankfurt am Main
> Germany
>
> T: +49 69 96769 278
> F: +49 69 96769 327
> Email: sauer.mpih at gmail.com
> www.brain.mpg.de
> <pos_fields.png>_______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>
>
>
>
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>

-- 
Dipl.-Psych. Andreas Sauer
Max Planck Institute for Brain Research
Deutschordenstraße 46
60528 Frankfurt am Main
Germany

T: +49 69 96769 278
F: +49 69 96769 327
Email: sauer.mpih at gmail.com
www.brain.mpg.de
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20160620/d510e12c/attachment-0002.html>