[FieldTrip] A few questions about source reconstruction method, templates and .gii file

[yhhu2020 at ion.ac.cn]

Hi everyone,

 I am doing source reconstruction and I have a few questions about the standardized template and how to choose source reconstruction method.

 At first, I followed the tutorial, and I know the .pos field represents the dipole positions in each subject. Now I want to plot the results using ft_sourceplot, which means I need to replace the positions with the normalized positions of the template sourcemodel. My question is where I can find those positions, i.e. a NumSource*3 matrix, such that I can draw source activity on surface. I have come up with a way that I use recon-all to MNI152_T1_2009c.nii and run ft_postfreesurferscript.sh. The .gii file in $SUBJECT_DIR/MNI152/workbench folder contains .pos matrix. Taking MNI152 as "one of subjects" and draw all results on its surface. I don't whether this procedure is right. After all, the distance between vertices is different and the coordinate system for gii can not be converted to neuromag.If this procedure is not right, can you show me some examples of right rpocedure?




 If I want to export my statistics data which were from ft_sourcestatistics function, how can I save my surface source activity using ft_sourcewrite? I try run below codes but the saved gii files can not be opened by workbench. I want to draw my surface results on workbench because someone told me the plotting is pretty in workbench.

 cfg = [];

 cfg.filetype = 'gifti';

 cfg.filename = 'tmp_surf';

 cfg.parameter = 'stat';

 ft_sourcewrite(cfg, stat_surf)




 The last question is what kind of source reconstruction method I should use? I find insufficient answers from here. I mean there are four situations: time-course of Event-related Field/Potential, time-course of oscillatory activity, Event-related Field/Potential at certain point in time and oscillatory activity at certain point in time. Do you have any recommendations for algorithms on MEG signals?




Sincerely thanks,

Yiheng




 

 








-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20241105/61e3061e/attachment.htm>