[FieldTrip] ft_sourceanalysis & ft_sourceinterpolate
Akiko Ikkai
akiko at nyu.edu
Fri Jun 24 17:42:20 CEST 2011
Hi,
I'm working on beamformer with MEG data, and having trouble
making ft_sourceanalysis and/or ft_sourceinterpolate work. It appears that
anatomical MRI and source data do not align, although anatomical MRI
("mri_aligned") and the result of ft_prepare_singleshell ("vol_cm") are both
aligned with MEG helmet.
Is there any way I could verify that the output of ft_sourceanalysis is not
distorted (i.e. same space as input)? I'm using fieldtrip-20110525 version.
Thanks in advance! Akiko
Below is my code:
1. compute the common filter
cfg = [];
cfg.grad = data_common.grad; % data_common = baseline + left +
right
cfg.grid.xgrid = -15:0.5:15;
cfg.grid.ygrid = -15:0.5:15;
cfg.grid.zgrid = -15:0.5:15;
cfg.inwardshift = -2;
cfg.vol = vol_cm; % this is from segmented "mri_aligned"
cfg.channel = {'all'};
cfg.reducerank = 2;
cfg.frequency = 10;
cfg.method = 'dics';
cfg.projectnoise = 'yes';
cfg.keepfilter = 'yes';
cfg.feedback = 'no';
cfg.realfilter = 'yes';
cfg.lambda = '0.005%';
source_common = ft_sourceanalysis(cfg, freq_common)
2. compute the spatial filter for one of the conditions (using grid & filter
from the step 1)
cfg = [];
cfg.grad = data_common.grad;
cfg.grid.pos = source_common.pos; % using grid & filter from the
common filter
cfg.grid.filter = source_common.avg.filter;
cfg.vol = vol_cm;
cfg.channel = {'all'};
cfg.reducerank = 2;
cfg.frequency = 10;
cfg.method = 'dics';
cfg.projectnoise = 'yes';
cfg.keepfilter = 'no';
cfg.feedback = 'no';
cfg.realfilter = 'yes';
cfg.lambda = '0.005%';
source_left = ft_sourceanalysis(cfg, freq_stiml);
source_left.unit = 'cm';
source_left.dim = source_common.dim;
3. calculate difference
source_left_bl = source_left;
source_left_bl.avg.pow = source_left_bl.avg.pow ./ source_common.avg.pow;
4. interpolate source to anatomical MRI space
cfg = [];
cfg.method = 'linear';
source_left_bl_int = ft_sourceinterpolate(cfg, source_left_bl,mri_aligned);
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20110624/34ab7428/attachment.html>
More information about the fieldtrip
mailing list