Hi Doug,
I've found a solution to the problem: I converted T1.mgz and FLAIR.mgz to nii.gz and then applied flirt directly with the same options as in flsregister called by bbregister, so
flirt -cost corratio -dof 6 -searchx -90 90 -searchy -90 90 \ -searchz -90 90 -in FLAIR.nii.gz -ref T1.nii.gz \ -out FLAIR_to_T1.nii.gz
This perfectly matched the T1.
Cheers, Ed
On 16 Jul 2012, at 18:00, freesurfer-request@nmr.mgh.harvard.edu wrote:
Message: 18 Date: Mon, 16 Jul 2012 11:32:06 -0400 From: Douglas N Greve greve@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] How to register FLAIR to T1? To: freesurfer@nmr.mgh.harvard.edu Message-ID: 500433F6.8010102@nmr.mgh.harvard.edu Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Often with FLAIR there is not enough gray-white contrast to run the registration. Try using the pial surface instead, with "-surf pial --t1" This tells it to use the pial surface as a boundary. The --t1 is needed because the intensity should go from bright to dark. Let me know if that does not fix the problem. doug
On 07/10/2012 11:58 AM, Bruce Fischl wrote:
glad to hear it :). You may need to repost (and possibly upload your data) On Tue, 10 Jul 2012, Ed Gronenschild wrote:
Hi Bruce,
It is indeed from the same scan session. I'll wait patiently, let Doug enjoy his vacation before he gets bothered again by all our questions.
Cheers, Ed
On 10 Jul 2012, at 16:57, Bruce Fischl wrote:
Hi Ed,
sorry, you'll have to wait until Doug is back from vacation. Is this from the same session? I've always had it work on flair images using one of the types of initialization.
cheers Bruce
On Tue, 10 Jul 2012, Ed Gronenschild wrote:
Hi Bruce,
Unfortunately, this has no effect.
Ed
On 10 Jul 2012, at 16:40, Bruce Fischl wrote:
Hi Ed
you are using -t2, right? You can also try -init-header
cheers Bruce On Tue, 10 Jul 2012, Ed Gronenschild wrote:
> Hi Bruce, > > It is slightly better but still displacements > in all directions and rotations too exist. > > Ed > > On 10 Jul 2012, at 16:13, Bruce Fischl wrote: > >> try -init-spm >> >> cheers >> Bruce >> On Tue, 10 Jul 2012, Ed Gronenschild wrote: >> >>> Hi, >>> >>> I would like to register FLAIR to T1. >>> Specifications FLAIR: >>> axial acquisition, 24 slices, >>> pixel size = 0.45 * 0.45 mm^2 >>> slice thickness = 5.50 mm (no gap). >>> >>> The slices run from the top of the brain >>> to just below the eyes. This is contrary >>> to the T1 scan which covers the full head >>> down to the neck. >>> I have tried bbregiser as follows: >>> >>> bbregister --s 001 --mov 001_FLAIR.mgz \ >>> --reg 001_register.dat --init-fsl --t1 --tol1d 1e-3 >>> >>> and inspected the result with >>> >>> tkregister2 --mov 001_FLAIR.mgz \ >>> --reg 001_register.dat --surf >>> >>> The result looks bad, it is displaced and rotated >>> by about 20 degrees with respect to T1. >>> >>> It makes hardly any difference if instead of --t1 >>> --t2 as contrast setting was used. >>> >>> Is there a better way to do this registration? >>> For instance, should I crop T1 to about the >>> same space as covered by FLAIR? >>> >>> (FreeSurfer v4.5.0, MacOSX 10.5) >>> >>> Cheers, >>> Ed >>> _______________________________________________ >>> 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. >> >
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Does fslregister work by itself?
On 7/17/12 10:39 AM, Ed Gronenschild wrote:
Hi Doug,
I've found a solution to the problem: I converted T1.mgz and FLAIR.mgz to nii.gz and then applied flirt directly with the same options as in flsregister called by bbregister, so
flirt -cost corratio -dof 6 -searchx -90 90 -searchy -90 90 \ -searchz -90 90 -in FLAIR.nii.gz -ref T1.nii.gz \ -out FLAIR_to_T1.nii.gz
This perfectly matched the T1.
Cheers, Ed
On 16 Jul 2012, at 18:00, freesurfer-request@nmr.mgh.harvard.edu mailto:freesurfer-request@nmr.mgh.harvard.edu wrote:
Message: 18
Date: Mon, 16 Jul 2012 11:32:06 -0400
From: Douglas N Greve <greve@nmr.mgh.harvard.edu mailto:greve@nmr.mgh.harvard.edu>
Subject: Re: [Freesurfer] How to register FLAIR to T1?
To: freesurfer@nmr.mgh.harvard.edu mailto:freesurfer@nmr.mgh.harvard.edu
Message-ID: <500433F6.8010102@nmr.mgh.harvard.edu mailto:500433F6.8010102@nmr.mgh.harvard.edu>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Often with FLAIR there is not enough gray-white contrast to run the
registration. Try using the pial surface instead, with "-surf pial --t1"
This tells it to use the pial surface as a boundary. The --t1 is needed
because the intensity should go from bright to dark. Let me know if that
does not fix the problem.
doug
On 07/10/2012 11:58 AM, Bruce Fischl wrote:
glad to hear it :). You may need to repost (and possibly upload your
data)
On Tue, 10 Jul 2012, Ed Gronenschild wrote:
Hi Bruce,
It is indeed from the same scan session.
I'll wait patiently, let Doug enjoy his vacation before
he gets bothered again by all our questions.
Cheers,
Ed
On 10 Jul 2012, at 16:57, Bruce Fischl wrote:
Hi Ed,
sorry, you'll have to wait until Doug is back from vacation. Is this from
the same session? I've always had it work on flair images using one of the
types of initialization.
cheers
Bruce
On Tue, 10 Jul 2012, Ed Gronenschild wrote:
Hi Bruce,
Unfortunately, this has no effect.
Ed
On 10 Jul 2012, at 16:40, Bruce Fischl wrote:
> Hi Ed > > > you are using -t2, right? You can also try -init-header > > > cheers > > Bruce > > On Tue, 10 Jul 2012, > > Ed Gronenschild wrote: > > >> Hi Bruce, >> >> >> It is slightly better but still displacements >> >> in all directions and rotations too exist. >> >> >> Ed >> >> >> On 10 Jul 2012, at 16:13, Bruce Fischl wrote: >> >> >>> try -init-spm >>> >>> >>> cheers >>> >>> Bruce >>> >>> On Tue, 10 Jul 2012, Ed Gronenschild wrote: >>> >>> >>>> Hi, >>>> >>>> >>>> I would like to register FLAIR to T1. >>>> >>>> Specifications FLAIR: >>>> >>>> axial acquisition, 24 slices, >>>> >>>> pixel size = 0.45 * 0.45 mm^2 >>>> >>>> slice thickness = 5.50 mm (no gap). >>>> >>>> >>>> The slices run from the top of the brain >>>> >>>> to just below the eyes. This is contrary >>>> >>>> to the T1 scan which covers the full head >>>> >>>> down to the neck. >>>> >>>> I have tried bbregiser as follows: >>>> >>>> >>>> bbregister --s 001 --mov 001_FLAIR.mgz \ >>>> >>>> --reg 001_register.dat --init-fsl --t1 --tol1d 1e-3 >>>> >>>> >>>> and inspected the result with >>>> >>>> >>>> tkregister2 --mov 001_FLAIR.mgz \ >>>> >>>> --reg 001_register.dat --surf >>>> >>>> >>>> The result looks bad, it is displaced and rotated >>>> >>>> by about 20 degrees with respect to T1. >>>> >>>> >>>> It makes hardly any difference if instead of --t1 >>>> >>>> --t2 as contrast setting was used. >>>> >>>> >>>> Is there a better way to do this registration? >>>> >>>> For instance, should I crop T1 to about the >>>> >>>> same space as covered by FLAIR? >>>> >>>> >>>> (FreeSurfer v4.5.0, MacOSX 10.5) >>>> >>>> >>>> Cheers, >>>> >>>> Ed >>>> >>>> _______________________________________________ >>>> >>>> Freesurfer mailing list >>>> >>>> Freesurfer@nmr.mgh.harvard.edu >>>> mailto: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. >>> >>> >>
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu mailto:Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Dear Freesurfer Team:
I would like some information on FWHM.dat folder glmdir and its relationship with mri_glmfit-sim. It happens that when I use mri_glmfit-sim with --cache-in can't find the data for fwhm 51 - which is logical because they exist only to 1..30, but my data were smoothed previously for fwhm 15 (pial_lgi). That interpretation would take this outcome? Something could go wrong when he observed these differences between the smoothing of my data and the reported in fwhm.dat
thanks in advance
I think that the LGI is a very smooth measure, so even if you only smoothed by 15mm, there is a lot of smoothness in the unsmoothed data. This is what fwhm.dat is picking up. What happens if you do not smooth the data? What fwhm do you get? doug
On 07/18/2012 08:02 AM, jm wrote:
Dear Freesurfer Team:
I would like some information on FWHM.dat folder glmdir and its relationship with mri_glmfit-sim. It happens that when I use mri_glmfit-sim with --cache-in can't find the data for fwhm 51 - which is logical because they exist only to 1..30, but my data were smoothed previously for fwhm 15 (pial_lgi). That interpretation would take this outcome? Something could go wrong when he observed these differences between the smoothing of my data and the reported in fwhm.dat
thanks in advance
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
On 07/18/2012 11:14 AM, Douglas N Greve wrote:
I think that the LGI is a very smooth measure, so even if you only smoothed by 15mm, there is a lot of smoothness in the unsmoothed data. This is what fwhm.dat is picking up. What happens if you do not smooth the data? What fwhm do you get? doug
On 07/18/2012 08:02 AM, jm wrote:
Dear Freesurfer Team:
I would like some information on FWHM.dat folder glmdir and its relationship with mri_glmfit-sim. It happens that when I use mri_glmfit-sim with --cache-in can't find the data for fwhm 51 - which is logical because they exist only to 1..30, but my data were smoothed previously for fwhm 15 (pial_lgi). That interpretation would take this outcome? Something could go wrong when he observed these differences between the smoothing of my data and the reported in fwhm.dat
thanks in advance
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
if I not smooth lgi data I get fwhm=13.2 into fwhm.dat
Hi Ed, the problem appears to be with the initialization. For some reason, fslregister is not initializing this data well. I don't know why the call directly to flirt works (and I've verified here that it does). When I initialize by hand or with SPM, bbregister works fine with the standard arguments. I'll need to dig into the flirt issue more. doug
On 07/17/2012 10:39 AM, Ed Gronenschild wrote:
Hi Doug,
I've found a solution to the problem: I converted T1.mgz and FLAIR.mgz to nii.gz and then applied flirt directly with the same options as in flsregister called by bbregister, so
flirt -cost corratio -dof 6 -searchx -90 90 -searchy -90 90 \ -searchz -90 90 -in FLAIR.nii.gz -ref T1.nii.gz \ -out FLAIR_to_T1.nii.gz
This perfectly matched the T1.
Cheers, Ed
On 16 Jul 2012, at 18:00, freesurfer-request@nmr.mgh.harvard.edu mailto:freesurfer-request@nmr.mgh.harvard.edu wrote:
Message: 18
Date: Mon, 16 Jul 2012 11:32:06 -0400
From: Douglas N Greve <greve@nmr.mgh.harvard.edu mailto:greve@nmr.mgh.harvard.edu>
Subject: Re: [Freesurfer] How to register FLAIR to T1?
To: freesurfer@nmr.mgh.harvard.edu mailto:freesurfer@nmr.mgh.harvard.edu
Message-ID: <500433F6.8010102@nmr.mgh.harvard.edu mailto:500433F6.8010102@nmr.mgh.harvard.edu>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Often with FLAIR there is not enough gray-white contrast to run the
registration. Try using the pial surface instead, with "-surf pial --t1"
This tells it to use the pial surface as a boundary. The --t1 is needed
because the intensity should go from bright to dark. Let me know if that
does not fix the problem.
doug
On 07/10/2012 11:58 AM, Bruce Fischl wrote:
glad to hear it :). You may need to repost (and possibly upload your
data)
On Tue, 10 Jul 2012, Ed Gronenschild wrote:
Hi Bruce,
It is indeed from the same scan session.
I'll wait patiently, let Doug enjoy his vacation before
he gets bothered again by all our questions.
Cheers,
Ed
On 10 Jul 2012, at 16:57, Bruce Fischl wrote:
Hi Ed,
sorry, you'll have to wait until Doug is back from vacation. Is this from
the same session? I've always had it work on flair images using one of the
types of initialization.
cheers
Bruce
On Tue, 10 Jul 2012, Ed Gronenschild wrote:
Hi Bruce,
Unfortunately, this has no effect.
Ed
On 10 Jul 2012, at 16:40, Bruce Fischl wrote:
> Hi Ed > > > you are using -t2, right? You can also try -init-header > > > cheers > > Bruce > > On Tue, 10 Jul 2012, > > Ed Gronenschild wrote: > > >> Hi Bruce, >> >> >> It is slightly better but still displacements >> >> in all directions and rotations too exist. >> >> >> Ed >> >> >> On 10 Jul 2012, at 16:13, Bruce Fischl wrote: >> >> >>> try -init-spm >>> >>> >>> cheers >>> >>> Bruce >>> >>> On Tue, 10 Jul 2012, Ed Gronenschild wrote: >>> >>> >>>> Hi, >>>> >>>> >>>> I would like to register FLAIR to T1. >>>> >>>> Specifications FLAIR: >>>> >>>> axial acquisition, 24 slices, >>>> >>>> pixel size = 0.45 * 0.45 mm^2 >>>> >>>> slice thickness = 5.50 mm (no gap). >>>> >>>> >>>> The slices run from the top of the brain >>>> >>>> to just below the eyes. This is contrary >>>> >>>> to the T1 scan which covers the full head >>>> >>>> down to the neck. >>>> >>>> I have tried bbregiser as follows: >>>> >>>> >>>> bbregister --s 001 --mov 001_FLAIR.mgz \ >>>> >>>> --reg 001_register.dat --init-fsl --t1 --tol1d 1e-3 >>>> >>>> >>>> and inspected the result with >>>> >>>> >>>> tkregister2 --mov 001_FLAIR.mgz \ >>>> >>>> --reg 001_register.dat --surf >>>> >>>> >>>> The result looks bad, it is displaced and rotated >>>> >>>> by about 20 degrees with respect to T1. >>>> >>>> >>>> It makes hardly any difference if instead of --t1 >>>> >>>> --t2 as contrast setting was used. >>>> >>>> >>>> Is there a better way to do this registration? >>>> >>>> For instance, should I crop T1 to about the >>>> >>>> same space as covered by FLAIR? >>>> >>>> >>>> (FreeSurfer v4.5.0, MacOSX 10.5) >>>> >>>> >>>> Cheers, >>>> >>>> Ed >>>> >>>> _______________________________________________ >>>> >>>> Freesurfer mailing list >>>> >>>> Freesurfer@nmr.mgh.harvard.edu >>>> mailto: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. >>> >>> >>
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu mailto:Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu
-
Douglas Greve -
Douglas N Greve
-
Ed Gronenschild -
jm