I wasn't sure about the exact hacking process and just naively commented
out lines 133-167 in the FreeSurfer script, so that I started with
recon-all -subjid $SubjectID -sd $SubjectDIR -autorecon2 -nosmooth2
-noinflate2 -nocurvstats -nosegstats -openmp ${num_cores}
${seed_cmd_appendix} (with brainmask and wm being edited)
It ran without problems, but stopped after saying
"mkdir: ribbon.postT2.pass1: File exists" in the Intermediate Recon-all
Steps. So it was rewriting everything up to this point but for some reason
stopped there.
Nevertheless, seeing that surfaces seemed to have been regenerated, I tried
to run PostFreeSurfer and apart from the warnings ("annot file:
/Users/user/Desktop/003/T1w/003/label/lh.BA.annot MRISreadAnnotationIntoArray:
vertex index out of range: 146448 i=00190519, in_array_size=146066) it went
without error and my myelin map looks good now.
What do you think? Can I use it or should I change something else so that
the complete pipeline finishes?
Thanks!
Lisa
On 7 June 2017 at 19:53, Glasser, Matthew <[email protected]> wrote:
> Right she would want to comment out stuff that had already been done.
>
> Matt.
>
> From: "Harms, Michael" <[email protected]>
> Date: Wednesday, June 7, 2017 at 12:50 PM
> To: Matt Glasser <[email protected]>, Lisa Kramarenko <
> [email protected]>
>
> Cc: "[email protected]" <[email protected]>
> Subject: Re: [HCP-Users] values in myelin maps
>
>
> I haven’t thought this all through, but I think that one of the main
> things you’d need to hack is to be able to make use of an already existing
> FS output. Currently, I believe that re-running the pipeline will simply
> overwrite any existing FS files, as if you were running FS de-novo.
>
> cheers,
> -MH
>
> --
> Michael Harms, Ph.D.
> -----------------------------------------------------------
> Conte Center for the Neuroscience of Mental Disorders
> Washington University School of Medicine
> Department of Psychiatry, Box 8134
> 660 South Euclid Ave. Tel: 314-747-6173 <(314)%20747-6173>
> St. Louis, MO 63110 Email: [email protected]
>
> From: "Glasser, Matthew" <[email protected]>
> Date: Wednesday, June 7, 2017 at 11:55 AM
> To: Lisa Kramarenko <[email protected]>, Michael Harms <
> [email protected]>
> Cc: "[email protected]" <[email protected]>
> Subject: Re: [HCP-Users] values in myelin maps
>
> Hi Lisa,
>
> I was going to suggest that you hack the pipeline yourself if you are
> comfortable with that. You might try starting at line 167 of the
> FreeSurfer pipeline to incorporate the brainmask edits (the WM edits might
> not be needed then). As for the control points, I don’t know enough about
> that to advise you so you’ll need to consult the FreeSurfer documentation
> to see where to add them in.
>
> The main things we do to modify the recon-all script is to assist with
> brain masking, to fine tune the white surface to be placed based on high
> res data, and place the pial surface based on highres T1w and T2w data
> using the script modules.
>
> The white surface tuning happens in between –white and –smooth2 whereas
> the pial surface tuning happens between -pial and -surfvolume on this
> table: http://surfer.nmr.mgh.harvard.edu/fswiki/ReconAllTableStableV5.3
>
> Peace,
> Matt.
>
> From: Lisa Kramarenko <[email protected]>
> Date: Wednesday, June 7, 2017 at 11:16 AM
> To: "Harms, Michael" <[email protected]>
> Cc: Matt Glasser <[email protected]>, "[email protected]" <
> [email protected]>
> Subject: Re: [HCP-Users] values in myelin maps
>
> so that means if something went wrong with the skull strip and surfaces
> during the FreeSurfer Pipeline there is no other alternative than to
> exclude these subjects?
>
> On 7 June 2017 at 18:04, Harms, Michael <[email protected]> wrote:
>
>>
>> Unfortunately, the HCP Pipelines do not currently support rerunning
>> FreeSurfer after editing. It is on our list of “things to do” to add that
>> functionality.
>>
>> cheers,
>> -MH
>>
>> --
>> Michael Harms, Ph.D.
>> -----------------------------------------------------------
>> Conte Center for the Neuroscience of Mental Disorders
>> Washington University School of Medicine
>> Department of Psychiatry, Box 8134
>> 660 South Euclid Ave.Tel: 314-747-6173 <(314)%20747-6173>
>> St. Louis, MO 63110Email: [email protected]
>>
>> From: <[email protected]> on behalf of Lisa
>> Kramarenko <[email protected]>
>> Date: Wednesday, June 7, 2017 at 11:00 AM
>> To: "Glasser, Matthew" <[email protected]>, "
>> [email protected]" <[email protected]>
>>
>> Subject: Re: [HCP-Users] values in myelin maps
>>
>> Dear Matthew,
>>
>> I have now manually edited brainmask.gmz ans wm.gmz to remove dura that
>> was still mistaken for a surface. Now I am not sure from which step to
>> re-run these participants.
>> 1. Do I understand it correctly that after replacing brainmask.gmz and
>> wm.gmz in the <subjid>/mri I need to start the FreeSurfer Pipeline after
>> the step "# Generate brain mask" (meaning starting with the step "# Call
>> recon-all to run most of the "-autorecon2" stages, but turning off smooth2,
>> inflate2, curvstats, and segstats stages")?
>> 2. Do I need to use any additional flags in the autorecon after my
>> intervention?
>> 3. I also created some control points to fix missing white matter. Should
>> I put the control.dat file in the <subjid>/tmp folder? Will FreeSurfer
>> see the file automatically or should I add some flag for it?
>>
>> thanks a lot for your help!
>> Lisa
>>
>> On 30 May 2017 at 16:38, Lisa Kramarenko <[email protected]>
>> wrote:
>>
>>> thanks
>>>
>>> best,
>>> Lisa
>>>
>>> On 30 May 2017 at 16:37, Glasser, Matthew <[email protected]> wrote:
>>>
>>>> Yes.
>>>>
>>>> Peace,
>>>>
>>>> Matt.
>>>>
>>>> From: Lisa Kramarenko <[email protected]>
>>>> Date: Tuesday, May 30, 2017 at 9:31 AM
>>>>
>>>> To: Matt Glasser <[email protected]>
>>>> Cc: "[email protected]" <[email protected]>
>>>> Subject: Re: [HCP-Users] values in myelin maps
>>>>
>>>> do I find the files I should inspect in the
>>>> {subject_id}/T1w/{subject_id}/mri/ folder?
>>>>
>>>> On 30 May 2017 at 16:28, Lisa Kramarenko <[email protected]>
>>>> wrote:
>>>>
>>>>> ok. so basically just go through recon-all output and manually correct
>>>>> whatever might have gone wrong?
>>>>>
>>>>> best,
>>>>> Lisa
>>>>>
>>>>> On 30 May 2017 at 16:13, Glasser, Matthew <[email protected]> wrote:
>>>>>
>>>>>> I would look for locations in which the FreeSurfer aseg is clearly
>>>>>> labeling things outside of the brain as grey or white matter.
>>>>>>
>>>>>> Peace,
>>>>>>
>>>>>> Matt.
>>>>>>
>>>>>> From: Lisa Kramarenko <[email protected]>
>>>>>> Date: Tuesday, May 30, 2017 at 8:59 AM
>>>>>>
>>>>>> To: Matt Glasser <[email protected]>
>>>>>> Cc: "[email protected]" <[email protected]>
>>>>>> Subject: Re: [HCP-Users] values in myelin maps
>>>>>>
>>>>>> Alright, thanks for the quick reply! Should I look for some specific
>>>>>> problem with the surfaces which is to be fixed? I mean I imagine a lot of
>>>>>> things can go wrong with a surface, so that I need to know what exactly
>>>>>> to
>>>>>> fix? Or is it something more general like this:
>>>>>> http://sites.bu.edu/cnrlab/lab-resources/freesurfer-qu
>>>>>> ality-control-guide/freesurfer-quality-control-step-1-fix-pi
>>>>>> al-surface/ ?
>>>>>>
>>>>>> And does it look like it affect both pial and white matter surfaces,
>>>>>> so that both need to be fixed?
>>>>>>
>>>>>> Sorry for the naive questions and thanks again.
>>>>>>
>>>>>> On 30 May 2017 at 15:51, Glasser, Matthew <[email protected]> wrote:
>>>>>>
>>>>>>> Fat saturation reduces the intensity of the fat within the bone
>>>>>>> marrow, which reduces the chance that FreeSurfer will mistake this fat
>>>>>>> for
>>>>>>> white matter. You would need to have this on during acquisition. As
>>>>>>> for
>>>>>>> how to fix the surfaces after the fact, I would look at FreeSurfer’s
>>>>>>> documentation.
>>>>>>>
>>>>>>> Peace,
>>>>>>>
>>>>>>> Matt.
>>>>>>>
>>>>>>> From: Lisa Kramarenko <[email protected]>
>>>>>>> Date: Tuesday, May 30, 2017 at 8:49 AM
>>>>>>>
>>>>>>> To: Matt Glasser <[email protected]>
>>>>>>> Cc: "[email protected]" <[email protected]>
>>>>>>> Subject: Re: [HCP-Users] values in myelin maps
>>>>>>>
>>>>>>> Yes, this is data of the lab I am in. I assume that I don't have fat
>>>>>>> saturation as I have never heard about it before... If I understand
>>>>>>> correctly it can't be done post-hoc after acquisition and should have
>>>>>>> been
>>>>>>> done during scanning? Other patients from the same batch (acquired the
>>>>>>> same
>>>>>>> way) don't have this problem.
>>>>>>> What would be the way to fix the surfaces manually?
>>>>>>>
>>>>>>> Thanks a lot!
>>>>>>>
>>>>>>> On 30 May 2017 at 15:42, Glasser, Matthew <[email protected]>
>>>>>>> wrote:
>>>>>>>
>>>>>>>> Are these of your own data? Do you have fat sat on in your T1w?
>>>>>>>> This looks like most likely surface errors because of not using fat
>>>>>>>> sat and
>>>>>>>> you would need to exclude these subjects or fix the surfaces manually.
>>>>>>>>
>>>>>>>> Peace,
>>>>>>>>
>>>>>>>> Matt.
>>>>>>>>
>>>>>>>> From: Lisa Kramarenko <[email protected]>
>>>>>>>> Date: Tuesday, May 30, 2017 at 7:53 AM
>>>>>>>> To: Matt Glasser <[email protected]>
>>>>>>>> Cc: "[email protected]" <[email protected]>
>>>>>>>> Subject: Re: [HCP-Users] values in myelin maps
>>>>>>>>
>>>>>>>> Thanks!
>>>>>>>> Another question I have concerning myelin maps is the following:
>>>>>>>> when I was doing quality control I noticed that some of them look
>>>>>>>> rather
>>>>>>>> weird (see screenshot). Why are some areas grey? And I presume the fact
>>>>>>>> that the values in this grey area are negative, indicates that
>>>>>>>> something is
>>>>>>>> wrong... And are the black parts normal (they still have positive
>>>>>>>> values)?
>>>>>>>> Do you have an idea of what to do?
>>>>>>>>
>>>>>>>> Thanks a lot!
>>>>>>>>
>>>>>>>> Lisa
>>>>>>>>
>>>>>>>> On 29 May 2017 at 15:55, Glasser, Matthew <[email protected]>
>>>>>>>> wrote:
>>>>>>>>
>>>>>>>>> They are the ratio of the T1w/T2w images and are a relative
>>>>>>>>> measure of myelin content. See this publication for more details:
>>>>>>>>>
>>>>>>>>> http://www.jneurosci.org/content/31/32/11597.short
>>>>>>>>>
>>>>>>>>> Peace,
>>>>>>>>>
>>>>>>>>> Matt.
>>>>>>>>>
>>>>>>>>> From: <[email protected]> on behalf of Lisa
>>>>>>>>> Kramarenko <[email protected]>
>>>>>>>>> Date: Monday, May 29, 2017 at 5:29 AM
>>>>>>>>> To: "[email protected]" <[email protected]
>>>>>>>>> >
>>>>>>>>> Subject: [HCP-Users] values in myelin maps
>>>>>>>>>
>>>>>>>>> Hello,
>>>>>>>>>
>>>>>>>>> I am not sure what exactly are the values shown for the myelin
>>>>>>>>> maps. E.g. on the screenshot the values are 0.998 to 1.771. I
>>>>>>>>> understand
>>>>>>>>> that higher values mean higher myelination and vice versa, but what
>>>>>>>>> do they
>>>>>>>>> mean exactly?
>>>>>>>>> Thanks!
>>>>>>>>>
>>>>>>>>> Lisa
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>>> HCP-Users mailing list
>>>>>>>>> [email protected]
>>>>>>>>> http://lists.humanconnectome.org/mailman/listinfo/hcp-users
>>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>> _______________________________________________
>> HCP-Users mailing list
>> [email protected]
>> http://lists.humanconnectome.org/mailman/listinfo/hcp-users
>>
>>
>> ------------------------------
>>
>> The materials in this message are private and may contain Protected
>> Healthcare Information or other information of a sensitive nature. If you
>> are not the intended recipient, be advised that any unauthorized use,
>> disclosure, copying or the taking of any action in reliance on the contents
>> of this information is strictly prohibited. If you have received this email
>> in error, please immediately notify the sender via telephone or return mail.
>>
>
>
_______________________________________________
HCP-Users mailing list
[email protected]
http://lists.humanconnectome.org/mailman/listinfo/hcp-users
