[FieldTrip] Issue aligning volume conduction model with source model

Doshi, Chiran Chiran.Doshi at childrens.harvard.edu
Wed Mar 20 16:00:34 CET 2013 

Will do,
Thanks!!

Chiran Doshi, MS
MEG system technologist
Department of Neurology
Boston Childrens Hospital
Phone: 781-216-1136
Fax: 781-216-1172


________________________________________
From: fieldtrip-bounces at science.ru.nl [fieldtrip-bounces at science.ru.nl] on behalf of Hamid Mohseni [hamid.mohseni at eng.ox.ac.uk]
Sent: Tuesday, March 19, 2013 11:51 AM
To: FieldTrip discussion list
Subject: Re: [FieldTrip] Issue aligning volume conduction model with source     model

Hi,

I had this problem before, and I think it is something with freesurfer: If you plot the original MRI and its segmented volume in freesurfer, you may see that they are not in the same coordinate. I think the best way is to ask freesurfer people, we probably missing a small thing here.

Thanks


On 19 March 2013 14:49, Doshi, Chiran <Chiran.Doshi at childrens.harvard.edu<mailto:Chiran.Doshi at childrens.harvard.edu>> wrote:
Hi,
So I tried form the beginning and it fixed the issue. Any ideas why this might be hapenning in the first place?
The problem is I already have a freesurfer database with reconstructed surfaces. If there is any fix like a missing transformation that I can compute so I do not have to recon-all again for all subjects.


Chiran Doshi, MS
MEG system technologist
Department of Neurology
Boston Childrens Hospital
Phone: 781-216-1136
Fax: 781-216-1172


________________________________________
From: fieldtrip-bounces at science.ru.nl<mailto:fieldtrip-bounces at science.ru.nl> [fieldtrip-bounces at science.ru.nl<mailto:fieldtrip-bounces at science.ru.nl>] on behalf of Doshi, Chiran [Chiran.Doshi at childrens.harvard.edu<mailto:Chiran.Doshi at childrens.harvard.edu>]
Sent: Monday, March 18, 2013 9:49 AM
To: FieldTrip discussion list
Subject: Re: [FieldTrip] Issue aligning volume conduction model with    source  model

Thanks for the help,
I will try again from the beginning. I will let know if I am able to get the fix the issue.

Chiran Doshi, MS
MEG system technologist
Department of Neurology
Boston Childrens Hospital
Phone: 781-216-1136
Fax: 781-216-1172


________________________________________
From: fieldtrip-bounces at science.ru.nl<mailto:fieldtrip-bounces at science.ru.nl> [fieldtrip-bounces at science.ru.nl<mailto:fieldtrip-bounces at science.ru.nl>] on behalf of Hamid Mohseni [hamid.mohseni at eng.ox.ac.uk<mailto:hamid.mohseni at eng.ox.ac.uk>]
Sent: Saturday, March 16, 2013 12:47 PM
To: FieldTrip discussion list
Subject: Re: [FieldTrip] Issue aligning volume conduction model with source     model

If you want to do beam former source localization, there is no need to freesurfer and you can extract the brain shape points from the segmented MRI using FieldTrip itself and you will not have this problem.

However, if you want to do minimum norm source localisation, you may do a realignment before applying free surfer, please see this

http://fieldtrip.fcdonders.nl/tutorial/minimumnormestimate

On Mar 16, 2013 4:23 PM, "Doshi, Chiran" <Chiran.Doshi at childrens.harvard.edu<mailto:Chiran.Doshi at childrens.harvard.edu><mailto:Chiran.Doshi at childrens.harvard.edu<mailto:Chiran.Doshi at childrens.harvard.edu>>> wrote:
Hi Hamid,
Thanks for the reply. I see what you mean.
When I import the T1, there is a mri.transform already present in it. This is generated by the freesurfer. mri.tranform gets changed when I do the ft_volumerealign, (following right hand system) I mark rpa as r, lpa as l , nas as n and z in the positive z direction. I have vitamin pills at the fiducials ( some also at the forehead) that would help in marking the points.

I cannot figure out where am I messing up. Also I am using the src file in MNE and I do not have this issue.

Chiran Doshi, MS
MEG system technologist
Department of Neurology
Boston Childrens Hospital
Phone: 781-216-1136
Fax: 781-216-1172

________________________________
From: fieldtrip-bounces at science.ru.nl<mailto:fieldtrip-bounces at science.ru.nl><mailto:fieldtrip-bounces at science.ru.nl<mailto:fieldtrip-bounces at science.ru.nl>> [fieldtrip-bounces at science.ru.nl<mailto:fieldtrip-bounces at science.ru.nl><mailto:fieldtrip-bounces at science.ru.nl<mailto:fieldtrip-bounces at science.ru.nl>>] on behalf of Hamid Mohseni [hamid.mohseni at eng.ox.ac.uk<mailto:hamid.mohseni at eng.ox.ac.uk><mailto:hamid.mohseni at eng.ox.ac.uk<mailto:hamid.mohseni at eng.ox.ac.uk>>]
Sent: Saturday, March 16, 2013 11:28 AM
To: FieldTrip discussion list
Subject: Re: [FieldTrip] Issue aligning volume conduction model with source model

Hi,

I THINK there is a problem with your mri.transform matrix. Basically, without any pre-processing, when you apply this transformation to the headshape points 'bnd', it should match the original mri. I checked that and apparently it dosen't happen. How did you obtain this transformation?

Thanks



On 15 March 2013 22:07, Doshi, Chiran <Chiran.Doshi at childrens.harvard.edu<mailto:Chiran.Doshi at childrens.harvard.edu><mailto:Chiran.Doshi at childrens.harvard.edu<mailto:Chiran.Doshi at childrens.harvard.edu>>> wrote:
Hello FieldTrip users,

I am having trouble aligning volume conduction model with source model

The link to the data is https://www.dropbox.com/sh/7ejsvewemci8ixe/RpfP3F5ROi

Here are the steps I followed

% Read mri
mri = ft_read_mri('T1.mgz');

% Determine coordinate system. Neuromag selected +x as r, +y as a, +z as s and origin as i. At the end mri.coordsys was equal to 'neuromag'
mri = ft_determine_coordsys(mri);

% Align volume. Selected l, r, l and z in + z direction
cfg=[];
cfg.coordsys='neuromag';
mri = ft_volumerealign(cfg,mri);

%Volume segment
cfg=[];
cfg.coordsys='neuromag';
cfg.output = {'brain' 'scalp' 'skull' 'tpm'};
seg = ft_volumesegment(cfg, mri);

% Prepare head model. The model aligns well with sensors
cfg=[];
cfg.coordsys='neuromag';
cfg.method    = 'singleshell';
vol = ft_prepare_headmodel(cfg, seg);

% compute sourcespace
bnd = ft_read_headshape('ID_000000-ico-5-src.fif','format','mne_source','unit','mm');
T   = mri.transform*inv(mri.transformorig);
sourcespace = ft_transform_geometry(T, bnd);

% Plot
figure;hold on;
ft_plot_mesh(sourcespace, 'edgecolor', 'none'); camlight
ft_plot_vol(vol, 'facecolor', 'none');alpha 0.5;



These surfaces do not align. Any ideas?



Chiran Doshi, MS
MEG system technologist
Department of Neurology
Boston Childrens Hospital
Phone: 781-216-1136
Fax: 781-216-1172

_______________________________________________
fieldtrip mailing list
fieldtrip at donders.ru.nl<mailto:fieldtrip at donders.ru.nl><mailto:fieldtrip at donders.ru.nl<mailto:fieldtrip at donders.ru.nl>>
http://mailman.science.ru.nl/mailman/listinfo/fieldtrip



--
Hamid R. Mohseni, PhD
Post-Doctoral Research Assistant
Institute of Biomedical Engineering
University of Oxford, OX3 7DQ, UK
Tel: +44 (0) 1865 2 83826<tel:%2B44%20%280%29%201865%202%2083826><tel:%2B44%20%280%29%201865%202%2083826>

_______________________________________________
fieldtrip mailing list
fieldtrip at donders.ru.nl<mailto:fieldtrip at donders.ru.nl><mailto:fieldtrip at donders.ru.nl<mailto:fieldtrip at donders.ru.nl>>
http://mailman.science.ru.nl/mailman/listinfo/fieldtrip

_______________________________________________
fieldtrip mailing list
fieldtrip at donders.ru.nl<mailto:fieldtrip at donders.ru.nl>
http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
_______________________________________________
fieldtrip mailing list
fieldtrip at donders.ru.nl<mailto:fieldtrip at donders.ru.nl>
http://mailman.science.ru.nl/mailman/listinfo/fieldtrip



--
Hamid R. Mohseni, PhD
Post-Doctoral Research Assistant
Institute of Biomedical Engineering
University of Oxford, OX3 7DQ, UK
Tel: +44 (0) 1865 2 83826

More information about the fieldtrip mailing list