Hello all!
I got no answer on this email, thus I'm posting these questions again, hoe someone can help:
Hello!
About synthesising a volume using mri_volsynth with direction cosines and RAS from mrs header:
In the mrs header I have: - VOI position in the 3 views (sag, cor and tra) which are from the left upper corner of the voxel (these are the RAS coordinates that I need to feed into mri_volsynth, correct?) - position vectors [0] [1] [2] - row and columns vectors (but I think that these are not the ones used to get the direction cosines, is that right?) - VOIRotationInPlane - and I know that my voxel is 3x3x3 cm
How do I introduce these infos in mri_volsynth to have a 3x3x3 cube to coregister with the anatomical? Typing mri_volsynth --help I got lots of asked information and I don't know how to feed into it. Also, for defining the geometry the "center" voxel and the first voxel are needed. Shouldn't it be only the RAS of the corner of the voxel?
Also, all these infos are in the scanner space, and the anatomical image processed in FS is spacially transformed. What should I use to coregister the synthesized volume to the processed anatomical to get the same transformation in the 3x3x3 volume?
Can you give me an example of how this process is done?
Thank you very much!
Andreia
Hi Doug,
Ok. Thank you! :)
Andreia
Citando Douglas N Greve greve@nmr.mgh.harvard.edu:
Yes, that should work. You will need a registration matrix to give to mri_compute_volume_fractions. You can get this from tkregister2:
tkregister2 --mov mrs.vol.nii --s subject --regheader --reg register.dat
when you run this, hit the compare button to make sure that the mrs box falls on the part of the anatomical you think it should be. Then hit "Save" to save the registration.
doug
_andreia_@sapo.pt wrote:
Hi Doug,
This already sheds some light on the issue! I didn't know mri_volsynth, I'm a begginer in FS. Lets see if I got it right. I have the anatomical and the MRS data acquired in the same session, so I won't need the EPI. If I manage to create the MRS volume using the header info and mri_volsynth I only need to coregister my anatomical with the synthetised volume and then run mri_compute_volume_fractions through the synthesed volume and the identity matrix (?) (since my data are from the same session) to get the % of the different tissue types only inside the voxel. Of course I'll need more advice on this later, but one thing at a time.
Thanks! Andreia
Citando Douglas N Greve greve@nmr.mgh.harvard.edu:
Sorry, I don't know anyone who can help with that. As you note, people do it, so there must be people around who can extract the relevant information out of the header. The information you need are the direction cosines and RAS of the corner of the voxel. You can feed this into mri_volsynth.
To register to FS, you'll need to have a whole-brain volume collected during the same session. If you collected the anatomical analyzed in FS at the same time, then you don't need another. If you did not, you can literally use any whole brain acquisition, eg, a single TR of an EPI (less than 10 sec to acquire). You can then register the EPI to the anatomical with bbregister. This gives you a "register.dat". You can then concatenate that with the EPI-MRS registration (I can show you how). After that, you can run mri_compute_volume_fractions passing it the synthesized volume and the register.dat (or identity matrix if same session).
Another alternative is to collect an actual MRI volume that spans your MRS volume. You can then substitute that for the synthetic volume. Note that the actual pixel content of this volume is irrelevant -- it's just used to get geometry information out of it. This will be the easiest if you can do it.
doug
_andreia_@sapo.pt wrote:
Hi Doug,
Do you know who can help me with this? That's exactly what I think that I need to do... The thing is that, unfortunately, I can't find anyone who can describe how the creation of an MRI "volume" from the MRS voxel is done, although this approach is present in many many papers.
About the second part, how do I register the synthetized MRI "volume" to the FS anatomical? And how can I get the volumes of only that "volume"?
(Hopefully I'll find a way to do the first part...)
Thanks, Andreia
Citando Douglas N Greve greve@nmr.mgh.harvard.edu:
Hi Andreia, I've done this before using information in the spectroscopy header, then synthesizing an MRI "volume" with one voxel that covers the size of your spectroscopy voxel, then registering that to the FS anatomical, then computing the partial volume fractions from the anatomical. The hard part is the first step, which I can't really advise you on.
doug
_andreia_@sapo.pt wrote: > Dear list, > > I am performing single-voxel and multivoxel spectroscopy analysis and > I need to do partial volume correction. I found a thread on the list > of someone asking if it possible to achieve this using Freesurfer > (see > below), but there is no answer saying how it is done. My question is > how can I define the voxel (for single-voxel) and the chosen voxels > (for multivoxel) in the anatomical data set to use the > segmentation of > only that defined voxel? I'm using a Siemens TrioTim 3T. All > acquisitions are in the same session. The anatomical dataset is an > average of 2 MPRAGEs which I've already processed using FS, and the > spectra files are .rda. The header of the MRS data has info on > vectors, but I don't know how to use them. > > > > "Re: [Freesurfer] need info for beginner > > Bruce Fischl > Thu, 26 Apr 2007 11:17:01 -0700 > > Hi Pom, > > yes, people here are using it for spectroscopy and for partial volume > correction (also for voxel placement). > > cheers, > Bruce > > On Thu, 26 Apr 2007 [EMAIL PROTECTED] wrote: > > Hi, > > I am about to install FREE SURFER for my Spectroscopy work. > I want to use the software to perform GM/WM segmentation on my > single voxel data sets. > > I am wondering if any one on this mailing list is currently > doing this task. I would like to get comments. > > My study is on a GE 3T scanner. > > thank you, > > Pom" > > > > Any advice, please? > > Thank you! > > Andreia > > > -- > > Andreia Pereira, MSc student > Visual Neuroscience Laboratory > IBILI-Faculdade de Medicina > Azinhaga de Santa Comba > 3000-354 Coimbra Portugal > Phone +(351) 239480220 > Fax +(351) 239480280 > > ------------------------------------------------------------------------ > > > > _______________________________________________ > Freesurfer mailing list > Freesurfer@nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
--
Andreia Pereira, MSc student Visual Neuroscience Laboratory IBILI-Faculdade de Medicina Azinhaga de Santa Comba 3000-354 Coimbra Portugal Phone +(351) 239480220 Fax +(351) 239480280
_andreia_@sapo.pt wrote:
Hello all!
I got no answer on this email, thus I'm posting these questions again, hoe someone can help:
Hello!
About synthesising a volume using mri_volsynth with direction cosines and RAS from mrs header:
In the mrs header I have:
- VOI position in the 3 views (sag, cor and tra) which are from the
left upper corner of the voxel (these are the RAS coordinates that I need to feed into mri_volsynth, correct?)
- position vectors [0] [1] [2]
- row and columns vectors (but I think that these are not the ones
used to get the direction cosines, is that right?)
- VOIRotationInPlane
- and I know that my voxel is 3x3x3 cm
I have no idea what the format of these files is, so it may take some trial and error to figure it out.
How do I introduce these infos in mri_volsynth to have a 3x3x3 cube to coregister with the anatomical? Typing mri_volsynth --help I got lots of asked information and I don't know how to feed into it. Also, for defining the geometry the "center" voxel and the first voxel are needed. Shouldn't it be only the RAS of the corner of the voxel?
The first voxel is specified with --p0. The direction cosines with --cdircos, etc, The voxel size (3mm) with --vox-size. The dimension will be --dim 1 1 1 1
Also, all these infos are in the scanner space, and the anatomical image processed in FS is spacially transformed. What should I use to coregister the synthesized volume to the processed anatomical to get the same transformation in the 3x3x3 volume?
FreeSurfer (eg, mri_vol2vol, tkregsiter2, etc) all take a --regheader argument that aligns things based on scanner coords. The FreeSurfer volumes keep track of scanner space.
doug
Can you give me an example of how this process is done?
Thank you very much!
Andreia
Hi Doug,
Ok. Thank you! :)
Andreia
Citando Douglas N Greve greve@nmr.mgh.harvard.edu:
Yes, that should work. You will need a registration matrix to give to mri_compute_volume_fractions. You can get this from tkregister2:
tkregister2 --mov mrs.vol.nii --s subject --regheader --reg register.dat
when you run this, hit the compare button to make sure that the mrs box falls on the part of the anatomical you think it should be. Then hit "Save" to save the registration.
doug
_andreia_@sapo.pt wrote:
Hi Doug,
This already sheds some light on the issue! I didn't know mri_volsynth, I'm a begginer in FS. Lets see if I got it right. I have the anatomical and the MRS data acquired in the same session, so I won't need the EPI. If I manage to create the MRS volume using the header info and mri_volsynth I only need to coregister my anatomical with the synthetised volume and then run mri_compute_volume_fractions through the synthesed volume and the identity matrix (?) (since my data are from the same session) to get the % of the different tissue types only inside the voxel. Of course I'll need more advice on this later, but one thing at a time.
Thanks! Andreia
Citando Douglas N Greve greve@nmr.mgh.harvard.edu:
Sorry, I don't know anyone who can help with that. As you note, people do it, so there must be people around who can extract the relevant information out of the header. The information you need are the direction cosines and RAS of the corner of the voxel. You can feed this into mri_volsynth.
To register to FS, you'll need to have a whole-brain volume collected during the same session. If you collected the anatomical analyzed in FS at the same time, then you don't need another. If you did not, you can literally use any whole brain acquisition, eg, a single TR of an EPI (less than 10 sec to acquire). You can then register the EPI to the anatomical with bbregister. This gives you a "register.dat". You can then concatenate that with the EPI-MRS registration (I can show you how). After that, you can run mri_compute_volume_fractions passing it the synthesized volume and the register.dat (or identity matrix if same session).
Another alternative is to collect an actual MRI volume that spans your MRS volume. You can then substitute that for the synthetic volume. Note that the actual pixel content of this volume is irrelevant -- it's just used to get geometry information out of it. This will be the easiest if you can do it.
doug
_andreia_@sapo.pt wrote:
Hi Doug,
Do you know who can help me with this? That's exactly what I think that I need to do... The thing is that, unfortunately, I can't find anyone who can describe how the creation of an MRI "volume" from the MRS voxel is done, although this approach is present in many many papers.
About the second part, how do I register the synthetized MRI "volume" to the FS anatomical? And how can I get the volumes of only that "volume"?
(Hopefully I'll find a way to do the first part...)
Thanks, Andreia
Citando Douglas N Greve greve@nmr.mgh.harvard.edu:
> Hi Andreia, I've done this before using information in the > spectroscopy > header, then synthesizing an MRI "volume" with one voxel that > covers the > size of your spectroscopy voxel, then registering that to the FS > anatomical, then computing the partial volume fractions from the > anatomical. The hard part is the first step, which I can't really > advise > you on. > > doug > > _andreia_@sapo.pt wrote: > >> Dear list, >> >> I am performing single-voxel and multivoxel spectroscopy analysis and >> I need to do partial volume correction. I found a thread on the list >> of someone asking if it possible to achieve this using Freesurfer >> (see >> below), but there is no answer saying how it is done. My question is >> how can I define the voxel (for single-voxel) and the chosen voxels >> (for multivoxel) in the anatomical data set to use the >> segmentation of >> only that defined voxel? I'm using a Siemens TrioTim 3T. All >> acquisitions are in the same session. The anatomical dataset is an >> average of 2 MPRAGEs which I've already processed using FS, and the >> spectra files are .rda. The header of the MRS data has info on >> vectors, but I don't know how to use them. >> >> >> >> "Re: [Freesurfer] need info for beginner >> >> Bruce Fischl >> Thu, 26 Apr 2007 11:17:01 -0700 >> >> Hi Pom, >> >> yes, people here are using it for spectroscopy and for partial volume >> correction (also for voxel placement). >> >> cheers, >> Bruce >> >> On Thu, 26 Apr 2007 [EMAIL PROTECTED] wrote: >> >> Hi, >> >> I am about to install FREE SURFER for my Spectroscopy work. >> I want to use the software to perform GM/WM segmentation on my >> single voxel data sets. >> >> I am wondering if any one on this mailing list is currently >> doing this task. I would like to get comments. >> >> My study is on a GE 3T scanner. >> >> thank you, >> >> Pom" >> >> >> >> Any advice, please? >> >> Thank you! >> >> Andreia >> >> >> -- >> >> Andreia Pereira, MSc student >> Visual Neuroscience Laboratory >> IBILI-Faculdade de Medicina >> Azinhaga de Santa Comba >> 3000-354 Coimbra Portugal >> Phone +(351) 239480220 >> Fax +(351) 239480280 >> >> ------------------------------------------------------------------------ >> >> >> >> _______________________________________________ >> Freesurfer mailing list >> Freesurfer@nmr.mgh.harvard.edu >> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer >> > -- > Douglas N. Greve, Ph.D. > MGH-NMR Center > greve@nmr.mgh.harvard.edu > Phone Number: 617-724-2358 > Fax: 617-726-7422 > > Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting > FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html > > _______________________________________________ > Freesurfer mailing list > Freesurfer@nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > > The information in this e-mail is intended only for the person to > whom it is > addressed. If you believe this e-mail was sent to you in error and > the e-mail > contains patient information, please contact the Partners Compliance > HelpLine at > http://www.partners.org/complianceline . If the e-mail was sent to > you in error > but does not contain patient information, please contact the sender > and properly > dispose of the e-mail. > > >
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
--
Andreia Pereira, MSc student Visual Neuroscience Laboratory IBILI-Faculdade de Medicina Azinhaga de Santa Comba 3000-354 Coimbra Portugal Phone +(351) 239480220 Fax +(351) 239480280
freesurfer@nmr.mgh.harvard.edu
-
_andreia_＠sapo.pt -
Douglas N Greve