Re: [Freesurfer] PETSurfer individual atlas for subcortocal segmentation

2019-12-05 Thread Boris Rauchmann 
External Email - Use Caution

Did you already have the time to look at the logfile? Do you have any 
suggestions how to proceed?

Thank you!

> Am 02.12.2019 um 19:18 schrieb Boris Rauchmann :
> 
> 
> In this example tried it with only the subcortical segmentations from my 
> atlas. Please find the logfile attached. It gives me back: "tissue type is 
> not set" but I set it to 2 in the LUT.txt
> 
> In principle look the following commands right to you?
> 
> xcerebralseg --s 0120test --o apas+head+subcort_BN.mgz --m 
> BN_Atlas_subcotex.mgz --atlas BN_Atlas_subcortex.gca
> 
> gtmseg --s 0120test --head apas+head+subcort_BN.mgz --ctab subcort_BN_LUT.txt 
> --o gtmseg+subcort_BN.mgz
> 
> Ideally I would have a gtmseg with both, the subcortical and the cortical 
> structures, but only the subcortical would also be fine as long as I can get  
> mri_gtmpvc running on it. 
> 
> Thanks,
> Boris
> 
>> On Mon, Dec 2, 2019 at 5:31 PM Greve, Douglas N.,Ph.D. 
>>  wrote:
>> Can you send the log file for each of the gtmseg runs?
>> 
>>> On 11/26/2019 1:09 PM, Boris Rauchmann wrote:
>>> External Email - Use Caution
>>> 
>>> Thank you! I have a gca for subcortical  and two gcs (lh/rh) for cortical 
>>> structures. 
>>> I created an annot (rh/lh) and a mgz using mris_ca_label and mri_ca_label 
>>> for parcellation/segmentation stats. 
>>> 
>>> For the PET analysis I have the following problem:
>>> 
>>> If I use this command: gtmseg --s test --o test.mgz --ctab 
>>> /xyz/BN_Atlas_246_LUT.txt --head BN_Atlas_subcotex.mgz --ctx-annot 
>>> BN_Atlas.annot --ctab '/xyz/BN_Atlas_246_LUT.txt' 
>>> 
>>> It gives me the right regions for subcortical structures but it looks like 
>>> it uses the standard FS parcellation with my labels for the cortical 
>>> parcellations (only 93 cortical regions instead of 210). 
>>> 
>>> If I use gtmseg --s 0059test --o onlyhead.gtmseg_test.mgz --ctx-annot 
>>> BN_Atlas.annot --ctab '/xyz/BN_Atlas_246_LUT.txt' --no-xcerseg I get all my 
>>> 210 cortical parcellations but the standard FS subcortical segmentations. 
>>> 
>>> How can I use both in one gtmseg so that I can proceed with it doing my PET 
>>> analysis in PETSurfer? It is not totally clear for me what to merge using 
>>> xcerebralseg.
>>> 
>>> Thanks a lot!
>>> 
 On Mon, Nov 18, 2019 at 8:28 PM Greve, Douglas N.,Ph.D. 
  wrote:
 It gets the subcortical from apas+head.mgz which gets created along the 
 way by xcerebralseg. You can create your own with xcerebralseg by 
 specifying your volume as the mergevol. I think this will work, but I'm 
 not sure. I'm assuming you've used the GCA to create your own 
 subcortical seg for the given subject
 
 On 11/5/19 1:06 PM, Boris Rauchmann wrote:
 >
 > External Email - Use Caution
 >
 > I just realized that the above mentioned command (gtmseg --s XYZ --o 
 > BN.gtmseg.mgz --ctx-annot BN_Atlas.annot --ctab 
 > '/media/XYZ/BN_Atlas_freesurfer/BN_Atlas_246_LUT.txt' --no-xcerseg) 
 > gives me only the cortical segmentation. Is there any way to also 
 > include the subcortical segmentation based on my individual atlas? I 
 > also have an Atlas_subcortex.gca file available.
 >
 > Best,
 > Boris
 >
 > On Tue, Aug 13, 2019 at 5:10 PM Greve, Douglas N.,Ph.D. 
 > mailto:dgr...@mgh.harvard.edu>> wrote:
 >
 > There is no cut off for the minimum size. As it gets smaller, the PVC
 > noise amplification will become bigger (it also depends on the
 > shape as
 > well).
 >
 > I think the --no-xcerseg is the right way to go now
 >
 > On 8/13/19 11:00 AM, Boris Rauchmann wrote:
 > >
 > > External Email - Use Caution
 > >
 > > Thank you for your prompt answer - the command worked. This is the
 > > atlas mentioned: http://atlas.brainnetome.org/brainnetome.html
 > > What is approximately the smallest possible segment when using PVC?
 > > Also, does the exclusion of extracerebral structures harm? I
 > used that
 > > flag because it complained:
 > >
 > > gtmseg --s XYZ --o BN.gtmseg.mgz --ctx-annot BN_Atlas.annot --ctab
 > > '/media/XYZ/BN_Atlas_freesurfer/BN_Atlas_246_LUT.txt'
 > > ERROR: /media/subjects/XYZ/mri/apas+head.mgz exists. This is ok
 > > but you must indicate whether to use what is there (--no-xcerseg)
 > > or create a new one and overwrite what is there (--xcerseg)
 > > or specify your own headseg (--head)
 > >
 > > and did not want to override my apas+head.mgz
 > >
 > > Thanks,
 > > Boris
 > >
 > > On Tue, Aug 13, 2019 at 4:44 PM Greve, Douglas N.,Ph.D.
 > > mailto:dgr...@mgh.harvard.edu>
 > >>
 > wrote:
 > >
 > > I don't kno

Re: [Freesurfer] PETSurfer individual atlas for subcortocal segmentation

2019-12-06 Thread Greve, Douglas N.,Ph.D. 
Hi Boris, I'll try to get to it today. Unfortunately, I think this is going to 
require a fair amount of effort on my part. When I wrote this part of 
PETsurfer, I just did not make it super flexible.
doug


On 12/5/2019 10:17 PM, Boris Rauchmann wrote:

External Email - Use Caution

Did you already have the time to look at the logfile? Do you have any 
suggestions how to proceed?

Thank you!

Am 02.12.2019 um 19:18 schrieb Boris Rauchmann 
:


In this example tried it with only the subcortical segmentations from my atlas. 
Please find the logfile attached. It gives me back: "tissue type is not set" 
but I set it to 2 in the LUT.txt

In principle look the following commands right to you?

xcerebralseg --s 0120test --o apas+head+subcort_BN.mgz --m 
BN_Atlas_subcotex.mgz --atlas BN_Atlas_subcortex.gca

gtmseg --s 0120test --head apas+head+subcort_BN.mgz --ctab subcort_BN_LUT.txt 
--o gtmseg+subcort_BN.mgz

Ideally I would have a gtmseg with both, the subcortical and the cortical 
structures, but only the subcortical would also be fine as long as I can get  
mri_gtmpvc running on it.

Thanks,
Boris

On Mon, Dec 2, 2019 at 5:31 PM Greve, Douglas N.,Ph.D. 
mailto:dgr...@mgh.harvard.edu>> wrote:
Can you send the log file for each of the gtmseg runs?

On 11/26/2019 1:09 PM, Boris Rauchmann wrote:

External Email - Use Caution

Thank you! I have a gca for subcortical  and two gcs (lh/rh) for cortical 
structures.
I created an annot (rh/lh) and a mgz using mris_ca_label and mri_ca_label for 
parcellation/segmentation stats.

For the PET analysis I have the following problem:

If I use this command: gtmseg --s test --o test.mgz --ctab 
/xyz/BN_Atlas_246_LUT.txt --head BN_Atlas_subcotex.mgz --ctx-annot 
BN_Atlas.annot --ctab '/xyz/BN_Atlas_246_LUT.txt'

It gives me the right regions for subcortical structures but it looks like it 
uses the standard FS parcellation with my labels for the cortical parcellations 
(only 93 cortical regions instead of 210).

If I use gtmseg --s 0059test --o onlyhead.gtmseg_test.mgz --ctx-annot 
BN_Atlas.annot --ctab '/xyz/BN_Atlas_246_LUT.txt' --no-xcerseg I get all my 210 
cortical parcellations but the standard FS subcortical segmentations.

How can I use both in one gtmseg so that I can proceed with it doing my PET 
analysis in PETSurfer? It is not totally clear for me what to merge using 
xcerebralseg.

Thanks a lot!

On Mon, Nov 18, 2019 at 8:28 PM Greve, Douglas N.,Ph.D. 
mailto:dgr...@mgh.harvard.edu>> wrote:
It gets the subcortical from apas+head.mgz which gets created along the
way by xcerebralseg. You can create your own with xcerebralseg by
specifying your volume as the mergevol. I think this will work, but I'm
not sure. I'm assuming you've used the GCA to create your own
subcortical seg for the given subject

On 11/5/19 1:06 PM, Boris Rauchmann wrote:
>
> External Email - Use Caution
>
> I just realized that the above mentioned command (gtmseg --s XYZ --o
> BN.gtmseg.mgz --ctx-annot BN_Atlas.annot --ctab
> '/media/XYZ/BN_Atlas_freesurfer/BN_Atlas_246_LUT.txt' --no-xcerseg)
> gives me only the cortical segmentation. Is there any way to also
> include the subcortical segmentation based on my individual atlas? I
> also have an Atlas_subcortex.gca file available.
>
> Best,
> Boris
>
> On Tue, Aug 13, 2019 at 5:10 PM Greve, Douglas N.,Ph.D.
> mailto:dgr...@mgh.harvard.edu> 
> >> wrote:
>
> There is no cut off for the minimum size. As it gets smaller, the PVC
> noise amplification will become bigger (it also depends on the
> shape as
> well).
>
> I think the --no-xcerseg is the right way to go now
>
> On 8/13/19 11:00 AM, Boris Rauchmann wrote:
> >
> > External Email - Use Caution
> >
> > Thank you for your prompt answer - the command worked. This is the
> > atlas mentioned: http://atlas.brainnetome.org/brainnetome.html
> > What is approximately the smallest possible segment when using PVC?
> > Also, does the exclusion of extracerebral structures harm? I
> used that
> > flag because it complained:
> >
> > gtmseg --s XYZ --o BN.gtmseg.mgz --ctx-annot BN_Atlas.annot --ctab
> > '/media/XYZ/BN_Atlas_freesurfer/BN_Atlas_246_LUT.txt'
> > ERROR: /media/subjects/XYZ/mri/apas+head.mgz exists. This is ok
> > but you must indicate whether to use what is there (--no-xcerseg)
> > or create a new one and overwrite what is there (--xcerseg)
> > or specify your own headseg (--head)
> >
> > and did not want to override my apas+head.mgz
> >
> > Thanks,
> > Boris
> >
> > On Tue, Aug 13, 2019 at 4:44 PM Greve, Douglas N.,Ph.D.
> > mailto:dgr...@mgh.harvard.edu> 
> >
>  
>

Re: [Freesurfer] PETSurfer individual atlas for subcortocal segmentation

2019-12-09 Thread Boris Rauchmann 
External Email - Use Caution

Dear Doug,

thank you for your effort. Let me know if you have any updates on this.

Best,
Boris

On Fri, Dec 6, 2019 at 5:45 PM Greve, Douglas N.,Ph.D. <
dgr...@mgh.harvard.edu> wrote:

> Hi Boris, I'll try to get to it today. Unfortunately, I think this is
> going to require a fair amount of effort on my part. When I wrote this part
> of PETsurfer, I just did not make it super flexible.
> doug
>
>
> On 12/5/2019 10:17 PM, Boris Rauchmann wrote:
>
> External Email - Use Caution
> Did you already have the time to look at the logfile? Do you have any
> suggestions how to proceed?
>
> Thank you!
>
> Am 02.12.2019 um 19:18 schrieb Boris Rauchmann 
> :
>
> 
> In this example tried it with only the subcortical segmentations from my
> atlas. Please find the logfile attached. It gives me back: "tissue type is
> not set" but I set it to 2 in the LUT.txt
>
> In principle look the following commands right to you?
>
> xcerebralseg --s 0120test --o apas+head+subcort_BN.mgz --m
> BN_Atlas_subcotex.mgz --atlas BN_Atlas_subcortex.gca
>
> gtmseg --s 0120test --head apas+head+subcort_BN.mgz --ctab
> subcort_BN_LUT.txt --o gtmseg+subcort_BN.mgz
>
> Ideally I would have a gtmseg with both, the subcortical and the cortical
> structures, but only the subcortical would also be fine as long as I can
> get  mri_gtmpvc running on it.
>
> Thanks,
> Boris
>
> On Mon, Dec 2, 2019 at 5:31 PM Greve, Douglas N.,Ph.D. <
> dgr...@mgh.harvard.edu> wrote:
>
>> Can you send the log file for each of the gtmseg runs?
>>
>> On 11/26/2019 1:09 PM, Boris Rauchmann wrote:
>>
>> External Email - Use Caution
>> Thank you! I have a gca for subcortical  and two gcs (lh/rh) for cortical
>> structures.
>> I created an annot (rh/lh) and a mgz using mris_ca_label and mri_ca_label
>> for parcellation/segmentation stats.
>>
>> For the PET analysis I have the following problem:
>>
>> If I use this command: gtmseg --s test --o test.mgz --ctab
>> /xyz/BN_Atlas_246_LUT.txt --head BN_Atlas_subcotex.mgz --ctx-annot
>> BN_Atlas.annot --ctab '/xyz/BN_Atlas_246_LUT.txt'
>>
>> It gives me the right regions for subcortical structures but it looks
>> like it uses the standard FS parcellation with my labels for the cortical
>> parcellations (only 93 cortical regions instead of 210).
>>
>> If I use gtmseg --s 0059test --o onlyhead.gtmseg_test.mgz --ctx-annot
>> BN_Atlas.annot --ctab '/xyz/BN_Atlas_246_LUT.txt' --no-xcerseg I get all my
>> 210 cortical parcellations but the standard FS subcortical segmentations.
>>
>> How can I use both in one gtmseg so that I can proceed with it doing my
>> PET analysis in PETSurfer? It is not totally clear for me what to merge
>> using xcerebralseg.
>>
>> Thanks a lot!
>>
>> On Mon, Nov 18, 2019 at 8:28 PM Greve, Douglas N.,Ph.D. <
>> dgr...@mgh.harvard.edu> wrote:
>>
>>> It gets the subcortical from apas+head.mgz which gets created along the
>>> way by xcerebralseg. You can create your own with xcerebralseg by
>>> specifying your volume as the mergevol. I think this will work, but I'm
>>> not sure. I'm assuming you've used the GCA to create your own
>>> subcortical seg for the given subject
>>>
>>> On 11/5/19 1:06 PM, Boris Rauchmann wrote:
>>> >
>>> > External Email - Use Caution
>>> >
>>> > I just realized that the above mentioned command (gtmseg --s XYZ --o
>>> > BN.gtmseg.mgz --ctx-annot BN_Atlas.annot --ctab
>>> > '/media/XYZ/BN_Atlas_freesurfer/BN_Atlas_246_LUT.txt' --no-xcerseg)
>>> > gives me only the cortical segmentation. Is there any way to also
>>> > include the subcortical segmentation based on my individual atlas? I
>>> > also have an Atlas_subcortex.gca file available.
>>> >
>>> > Best,
>>> > Boris
>>> >
>>> > On Tue, Aug 13, 2019 at 5:10 PM Greve, Douglas N.,Ph.D.
>>> > mailto:dgr...@mgh.harvard.edu>> wrote:
>>> >
>>> > There is no cut off for the minimum size. As it gets smaller, the
>>> PVC
>>> > noise amplification will become bigger (it also depends on the
>>> > shape as
>>> > well).
>>> >
>>> > I think the --no-xcerseg is the right way to go now
>>> >
>>> > On 8/13/19 11:00 AM, Boris Rauchmann wrote:
>>> > >
>>> > > External Email - Use Caution
>>> > >
>>> > > Thank you for your prompt answer - the command worked. This is
>>> the
>>> > > atlas mentioned: http://atlas.brainnetome.org/brainnetome.html
>>> > > What is approximately the smallest possible segment when using
>>> PVC?
>>> > > Also, does the exclusion of extracerebral structures harm? I
>>> > used that
>>> > > flag because it complained:
>>> > >
>>> > > gtmseg --s XYZ --o BN.gtmseg.mgz --ctx-annot BN_Atlas.annot
>>> --ctab
>>> > > '/media/XYZ/BN_Atlas_freesurfer/BN_Atlas_246_LUT.txt'
>>> > > ERROR: /media/subjects/XYZ/mri/apas+head.mgz exists. This is ok
>>> > > but you must indicate whether to use what is there (--no-xcerseg)
>>> > > or create a new one and overwrite what is there (--xcerseg)
>>