[FieldTrip] source localization difficulties

jan-mathijs schoffelen jan.schoffelen at donders.ru.nl
Wed Sep 21 09:19:15 CEST 2011


Hi Elli,

> Firstly, and most problematic, examining the same MRI data using `ft_read_mri` + `ft_volumerealign` and the tkmedit tool that comes with FreeSurfer seems to show two completely different brains. I'm unable to locate landmarks in the FT browser that I can see in the tkmedit window, and vice versa. The data isn't simply flipped; it looks that each slice is different. Why would this be the case?

ft_read_mri reads in the raw MRI-file(s) that you specified (e.g. the DICOM-slices) and stacks them in the right order. When displaying the data in ft_volumerealign, no magic is performed and you will just see the data as it is stored in the 3D-volume. Nothing more, nothing less. Did you apply any additional processing steps before displaying it with freesurfer?
It is hard for me to judge whether something, and if so, what is going wrong here.

> 
> Secondly, in an attempt to coregister the MRI and MEG data, I used `ft_determine_coordsys`, which simply opened a window with a picture of three axes (see attached image "ft_determine_coordsys.png"). Shouldn't there be image data overlaid on the axes?
> 

There's no figure attached in the mail I got. I would indeed expect to see 3 axes and three intersecting orthogonal slices with anatomical data. That is, if the call to ft_determine_coordsys was ft_determine_coordsys([],mri). Note that this function is just intended to check whether the coordinate system attached to the input data (mri, sensor positions, volume conduction model of the head, source dipole positions etc.) is as expected, or to assign the coordinate system, if you provide some input as to how to interpret the 3 cardinal axes.
In order to check whether the co-registration went OK I would 
-co-register the MRI to the coordinate system in which the meg sensors are expressed
-segment the MRI (these previous 2 steps can be done in freesurfer 
-create a volume conduction model (let's call it 'vol') of the head (e.g. by calling ft_prepare_singleshell, or even better: call ft_prepare_headmodel with cfg.method = 'singleshell')
-check the co-registration:  ft_plot_sens(data.grad);hold on;ft_plot_vol(vol), this should show the brain surface and the sensors more or less properly aligned.

> Lastly, I've already segmented and coregistered the MRI and MEG data using FreeSurfer. Is there a way to simply load this data into FieldTrip and get right to plotting the data?
> 

Yes, as far as I know ft_read_mri should know how to deal with *.mgz volumes.

> Thanks!

You're welcome.

Jan-Mathijs

> 
> 
> --------------------
> Eliezer Kanal, Ph.D.
> Postdoctoral Fellow
> Center for the Neural Basis of Cognition
> Carnegie Mellon University
> 4400 Fifth Ave, Suite 110A
> Pittsburgh PA 15213
> P: 412-268-4115
> F: 412-268-5060
> 
> 
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip

Jan-Mathijs Schoffelen, MD PhD 
Donders Institute for Brain, Cognition and Behaviour, 
Centre for Cognitive Neuroimaging,
Radboud University Nijmegen, The Netherlands
J.Schoffelen at donders.ru.nl
Telephone: +31-24-3614793

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20110921/a57cf276/attachment.html>


More information about the fieldtrip mailing list