A new version of Freesurfer is now available for download from this page:
https://surfer.nmr.mgh.harvard.edu/fswiki/Download
The list of new features, bug fixes, and known issues is found on the ReleaseNotes page:
https://surfer.nmr.mgh.harvard.edu/fswiki/ReleaseNotes
Users are reminded that in general, subjects should be processed by a single version of Freesurfer (ie. in a group study, subjects should not have been processed by different versions of Freesurfer). Older versions of Freesurfer are available upon request.
The Freesurfer Team.
Dear Freesurfers,
I would like a copy of freesurfer-Linux-rh9-stable-pub-v3.0.5 since most of my data has been processed using that version. As per your suggestion, I have decided to stick to the same version of the program. I do have a question for you though: I am analyzing data from a longitudinal study with three different time points. I am nearing the end of the first time point. Would you still suggest that I stick to the same version of freesurfer for my two remaining time points?
thanks in advance,
Binyam
Binyam Nardos Cognitive Rehabilitation Research Group Washington University School of Medicine Box # 8505 4444 Forest Park Blvd St. Louis, MO 63108 Office: (314) 454-7795 Fax: (314) 286-1601
The contents of this e-mail message and any attachments are intended solely for the addressee(s) named in this message. This communication is intended to be and to remain confidential and may be subject to applicable physician/patient and/or work product privileges. If you are not the intended recipient of this message, or if this message has been addressed to you in error, please immediately alert the sender by reply e-mail and then delete this message and its attachments.
-----Original Message----- From: freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Nick Schmansky Sent: Wed 8/22/2007 8:01 AM To: Freesurfer Mailing List Subject: [Freesurfer] New Freesurfer v4.0.0
A new version of Freesurfer is now available for download from this page:
https://surfer.nmr.mgh.harvard.edu/fswiki/Download
The list of new features, bug fixes, and known issues is found on the ReleaseNotes page:
https://surfer.nmr.mgh.harvard.edu/fswiki/ReleaseNotes
Users are reminded that in general, subjects should be processed by a single version of Freesurfer (ie. in a group study, subjects should not have been processed by different versions of Freesurfer). Older versions of Freesurfer are available upon request.
The Freesurfer Team.
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Binyam,
The v3.0.5 rh9 build of freesurfer can be downloaded from here:
ftp://surfer.nmr.mgh.harvard.edu/pub/dist
For the remaining two time points, you should use the same freesurfer version as the first time point.
Nick
On Wed, 2007-08-22 at 13:41 -0500, Binyam Nardos wrote:
Dear Freesurfers,
I would like a copy of freesurfer-Linux-rh9-stable-pub-v3.0.5 since most of my data has been processed using that version. As per your suggestion, I have decided to stick to the same version of the program. I do have a question for you though: I am analyzing data from a longitudinal study with three different time points. I am nearing the end of the first time point. Would you still suggest that I stick to the same version of freesurfer for my two remaining time points?
thanks in advance,
Binyam
Binyam Nardos Cognitive Rehabilitation Research Group Washington University School of Medicine Box # 8505 4444 Forest Park Blvd St. Louis, MO 63108 Office: (314) 454-7795 Fax: (314) 286-1601
The contents of this e-mail message and any attachments are intended solely for the addressee(s) named in this message. This communication is intended to be and to remain confidential and may be subject to applicable physician/patient and/or work product privileges. If you are not the intended recipient of this message, or if this message has been addressed to you in error, please immediately alert the sender by reply e-mail and then delete this message and its attachments.
-----Original Message----- From: freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Nick Schmansky Sent: Wed 8/22/2007 8:01 AM To: Freesurfer Mailing List Subject: [Freesurfer] New Freesurfer v4.0.0
A new version of Freesurfer is now available for download from this page:
https://surfer.nmr.mgh.harvard.edu/fswiki/Download
The list of new features, bug fixes, and known issues is found on the ReleaseNotes page:
https://surfer.nmr.mgh.harvard.edu/fswiki/ReleaseNotes
Users are reminded that in general, subjects should be processed by a single version of Freesurfer (ie. in a group study, subjects should not have been processed by different versions of Freesurfer). Older versions of Freesurfer are available upon request.
The Freesurfer Team.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Binyam,
Bruce has advised me that you would be better off re-processing your existing subjects with v4.0, rather than processing those future time- points with v3.0.5. There are a few reasons: the main one is that v4.0 includes the -long flag for performing longitudinal processing (there is a section on this in recon-all --help). v3.0.5 does not have that flag, and is a bit of a hack to incorporate it into v3.0.5. Secondly, v4.0 should do a better job overall in processing than v3.0.5. Thirdly, for an ongoing study, you probably would want to make use of present and future tools in the v4.0 generation of code.
Nick
On Wed, 2007-08-22 at 14:52 -0400, Nick Schmansky wrote:
Binyam,
The v3.0.5 rh9 build of freesurfer can be downloaded from here:
ftp://surfer.nmr.mgh.harvard.edu/pub/dist
For the remaining two time points, you should use the same freesurfer version as the first time point.
Nick
On Wed, 2007-08-22 at 13:41 -0500, Binyam Nardos wrote:
Dear Freesurfers,
I would like a copy of freesurfer-Linux-rh9-stable-pub-v3.0.5 since most of my data has been processed using that version. As per your suggestion, I have decided to stick to the same version of the program. I do have a question for you though: I am analyzing data from a longitudinal study with three different time points. I am nearing the end of the first time point. Would you still suggest that I stick to the same version of freesurfer for my two remaining time points?
thanks in advance,
Binyam
Binyam Nardos Cognitive Rehabilitation Research Group Washington University School of Medicine Box # 8505 4444 Forest Park Blvd St. Louis, MO 63108 Office: (314) 454-7795 Fax: (314) 286-1601
The contents of this e-mail message and any attachments are intended solely for the addressee(s) named in this message. This communication is intended to be and to remain confidential and may be subject to applicable physician/patient and/or work product privileges. If you are not the intended recipient of this message, or if this message has been addressed to you in error, please immediately alert the sender by reply e-mail and then delete this message and its attachments.
-----Original Message----- From: freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Nick Schmansky Sent: Wed 8/22/2007 8:01 AM To: Freesurfer Mailing List Subject: [Freesurfer] New Freesurfer v4.0.0
A new version of Freesurfer is now available for download from this page:
https://surfer.nmr.mgh.harvard.edu/fswiki/Download
The list of new features, bug fixes, and known issues is found on the ReleaseNotes page:
https://surfer.nmr.mgh.harvard.edu/fswiki/ReleaseNotes
Users are reminded that in general, subjects should be processed by a single version of Freesurfer (ie. in a group study, subjects should not have been processed by different versions of Freesurfer). Older versions of Freesurfer are available upon request.
The Freesurfer Team.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Dear FreeSurfers,
To produce cortical gray matter ribbon masks for each annotation roi, we have used in the past the FreeSurfer executable mris_annot_to_segmentation, and then extracted our 3D masks from the resulting integer code image (in brains2).
The current release of mris_annot_to_segmentation works ok on FreeSurfer studies we ran, for example, on April 15, 2006; however, on FreeSurfer studies generated with a more modern version of FreeSurfer, for example, on March 1, 2007, mris_annot_to_segmentation dies with a SegFault and (in gdb) this sort of message:
Program received signal EXC_BAD_ACCESS, Could not access memory. Reason: KERN_INVALID_ADDRESS at address: 0x804d1000 0x00002df2 in ?? () (gdb) backtrace #0 0x00002df2 in ?? () #1 0x01807400 in ?? () Previous frame inner to this frame (corrupt stack?) (gdb)
Together with the advice from Nick reproduced below, this seems to me to mean I do not need an earlier version of mris_annot_to_segmentation since the current version is continuous with data generated last year; instead, I need to request that someone at FreeSurfer headquarters figure out how to compile mris_annot_to_segmentation for the contemporary file formats now in use for surf/lh.white, surf/lh.pial, label/lh.aparc.annot, etc., right?
Unless the best thing would be to convert these files to the older file format with something like mris_convert from two distinct releases of FreeSurfer? I think I see how to convert white and pial, but what converts aparc.annot to text and back to binary? And isn't my difficulty one that straddles a version before v.3.0.3 on 5/24/06? Can I even get the earlier version from January 2006?
I can ftp a tarball with one of our workups and a TESTME.csh script in it, but it is 234MB so I am hoping to finesse that if possible.
Thanks for any and all suggestions.
Greg Harris Iowa Psychiatry Brain Imaging Lab
On 8/22/07 8:01 AM, "Nick Schmansky" nicks@nmr.mgh.harvard.edu wrote:
A new version of Freesurfer is now available for download from this page:
https://surfer.nmr.mgh.harvard.edu/fswiki/Download
The list of new features, bug fixes, and known issues is found on the ReleaseNotes page:
https://surfer.nmr.mgh.harvard.edu/fswiki/ReleaseNotes
Users are reminded that in general, subjects should be processed by a single version of Freesurfer (ie. in a group study, subjects should not have been processed by different versions of Freesurfer). Older versions of Freesurfer are available upon request.
The Freesurfer Team.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Hi Nick and other surfers, I shot another email back yesterday, not sure if you received it. This is regarding a longitudinal study I'm working on. Most of data point one has been processed using version 3.05. I have already processed about 50 subjects to be exact and considering freesurfer's time consumption, it is very unfortunate that they have to be done over again, as per your and Bruce's suggestion due to version 4's suitability for longitudinal processing. But I guess it has to be done ... now I was wondering if the entire processing stream have to be repeated or is there a latter point that I can start processing the data from to accomodate the new changes implemented in v 4.0?
Appreciate your help,
Binyam
Binyam Nardos Cognitive Rehabilitation Research Group Washington University School of Medicine Box # 8505 4444 Forest Park Blvd St. Louis, MO 63108 Office: (314) 454-7795 Fax: (314) 286-1601
The contents of this e-mail message and any attachments are intended solely for the addressee(s) named in this message. This communication is intended to be and to remain confidential and may be subject to applicable physician/patient and/or work product privileges. If you are not the intended recipient of this message, or if this message has been addressed to you in error, please immediately alert the sender by reply e-mail and then delete this message and its attachments.
Hi Binyam,
you don't have to do them again if you don't want to, but 4.0 has longitudinal processing in it, and 3.05 doesn't. Note that reprocessing shouldn't require any additional manual interaction - you just rerun the automated steps. You will also find that processing the additional timepoints it *much* easier with the longitudinal stream.
cheers, Bruce On Thu, 23 Aug 2007, Binyam Nardos wrote:
Hi Nick and other surfers, I shot another email back yesterday, not sure if you received it. This is regarding a longitudinal study I'm working on. Most of data point one has been processed using version 3.05. I have already processed about 50 subjects to be exact and considering freesurfer's time consumption, it is very unfortunate that they have to be done over again, as per your and Bruce's suggestion due to version 4's suitability for longitudinal processing. But I guess it has to be done ... now I was wondering if the entire processing stream have to be repeated or is there a latter point that I can start processing the data from to accomodate the new changes implemented in v 4.0?
Appreciate your help,
Binyam
Binyam Nardos Cognitive Rehabilitation Research Group Washington University School of Medicine Box # 8505 4444 Forest Park Blvd St. Louis, MO 63108 Office: (314) 454-7795 Fax: (314) 286-1601
The contents of this e-mail message and any attachments are intended solely for the addressee(s) named in this message. This communication is intended to be and to remain confidential and may be subject to applicable physician/patient and/or work product privileges. If you are not the intended recipient of this message, or if this message has been addressed to you in error, please immediately alert the sender by reply e-mail and then delete this message and its attachments.
Nick and Freesurfer Team,
This is a followup to the message sent by Greg Harris yesterday. Greg has put together a tarball and shell script that demonstrates the problem.
wget http://www.psychiatry.uiowa.edu/users/hjohnson/ftp/tempFScase.tgz tar -xzvf /iplweb/html/users/hjohnson/ftp/tempFScase.tgz cd tempFScase/ csh TESTME.csh
I am hoping that if this were run through a debugger that the problem is going to be quite easy to fix. Will you please let us know if there is anything that we can do to help address this issue?
Thanks, Hans
Hi Hans,
I fixed this last night, Nick can give you a new version if you tell him what hardware/software platform you are using. However, looking at the data it seems that you already stripped the cerebellum from the orig.mgz volume. This will mess everything up, as the subsequent atlas alignment will try to match a (non-existent) cerebellum in the data and the aseg will do something strange. Are you stripping it using something else before putting the data through recon-all?
cheers, Bruce
On Thu, 23 Aug 2007, Hans Johnson wrote:
Nick and Freesurfer Team,
This is a followup to the message sent by Greg Harris yesterday. Greg has put together a tarball and shell script that demonstrates the problem.
wget http://www.psychiatry.uiowa.edu/users/hjohnson/ftp/tempFScase.tgz tar -xzvf /iplweb/html/users/hjohnson/ftp/tempFScase.tgz cd tempFScase/ csh TESTME.csh
I am hoping that if this were run through a debugger that the problem is going to be quite easy to fix. Will you please let us know if there is anything that we can do to help address this issue?
Thanks, Hans
Bruce,
Thank you for the quick response. I'll need to speak to the developer who created the data processing scripts that generated the data sent to you before I can give an informed answer.
I've already sent Nick an e-mail requesting either the "32bit linux" or "64bit linux" version of the code.
Again, thank you for the quick response.
Hans
Hans,
I've posted the new mris_annot_to_segmentation for each of the linux platforms here:
ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/fsdev/nicks/stable
copy the file to your $FREESURFER_HOME/bin, then type: chmod a+x $FREESURFER_HOME/bin/mris_annot_to_segmentation
This fixed binary will be in the next release (v4.0.1).
Nick
On Fri, 2007-08-24 at 10:00 -0500, Hans Johnson wrote:
Bruce,
Thank you for the quick response. I'll need to speak to the developer who created the data processing scripts that generated the data sent to you before I can give an informed answer.
I've already sent Nick an e-mail requesting either the "32bit linux" or "64bit linux" version of the code.
Again, thank you for the quick response.
Hans
Nick and Bruce,
THANKS. No segfault now! We will investigate the processing steps in the next few days, and get back to you with our findings.
Hans
Hi Hans,
have you tried using mri_aparc2aseg instead of mris_annot_to_segmentation?
doug
Bruce Fischl wrote:
Hi Hans,
I fixed this last night, Nick can give you a new version if you tell him what hardware/software platform you are using. However, looking at the data it seems that you already stripped the cerebellum from the orig.mgz volume. This will mess everything up, as the subsequent atlas alignment will try to match a (non-existent) cerebellum in the data and the aseg will do something strange. Are you stripping it using something else before putting the data through recon-all?
cheers, Bruce
On Thu, 23 Aug 2007, Hans Johnson wrote:
Nick and Freesurfer Team,
This is a followup to the message sent by Greg Harris yesterday. Greg has put together a tarball and shell script that demonstrates the problem.
wget http://www.psychiatry.uiowa.edu/users/hjohnson/ftp/tempFScase.tgz tar -xzvf /iplweb/html/users/hjohnson/ftp/tempFScase.tgz cd tempFScase/ csh TESTME.csh
I am hoping that if this were run through a debugger that the problem is going to be quite easy to fix. Will you please let us know if there is anything that we can do to help address this issue?
Thanks, Hans
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Hi Hans,
I think mris_annot_to_segmentation is deprecated. Did I tell you to use it at some point?
Bruce On Fri, 24 Aug 2007, Doug Greve wrote:
Hi Hans,
have you tried using mri_aparc2aseg instead of mris_annot_to_segmentation?
doug
Bruce Fischl wrote:
Hi Hans,
I fixed this last night, Nick can give you a new version if you tell him what hardware/software platform you are using. However, looking at the data it seems that you already stripped the cerebellum from the orig.mgz volume. This will mess everything up, as the subsequent atlas alignment will try to match a (non-existent) cerebellum in the data and the aseg will do something strange. Are you stripping it using something else before putting the data through recon-all?
cheers, Bruce
On Thu, 23 Aug 2007, Hans Johnson wrote:
Nick and Freesurfer Team,
This is a followup to the message sent by Greg Harris yesterday. Greg has put together a tarball and shell script that demonstrates the problem.
wget http://www.psychiatry.uiowa.edu/users/hjohnson/ftp/tempFScase.tgz tar -xzvf /iplweb/html/users/hjohnson/ftp/tempFScase.tgz cd tempFScase/ csh TESTME.csh
I am hoping that if this were run through a debugger that the problem is going to be quite easy to fix. Will you please let us know if there is anything that we can do to help address this issue?
Thanks, Hans
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Bruce,
I don't know who advised us to use mris_annot_to_segmentation, but based on the helpful comments from the freesurfer team, we have moved to using mri_aparc2aseg instead.
Thanks Doug!
Hans
Bruce,
Thank you for your comments and help. We are aware that this particular data set does not have the cerebellum. This is not part of our standard protocol, and we are not going to publish on this data. This is data that was given to us "pre-stripped" of both skull and cerebellum. We have been playing with it to see if we can resurrect any useful information from it.
Our standard processing with freesurfer does include the cerebellum.
Regards, Hans
ok, I'm surprised it worked as well as it did.
cheers, Bruce On Tue, 4 Sep 2007, Hans Johnson wrote:
Bruce,
Thank you for your comments and help. We are aware that this particular data set does not have the cerebellum. This is not part of our standard protocol, and we are not going to publish on this data. This is data that was given to us "pre-stripped" of both skull and cerebellum. We have been playing with it to see if we can resurrect any useful information from it.
Our standard processing with freesurfer does include the cerebellum.
Regards, Hans
freesurfer@nmr.mgh.harvard.edu
-
Binyam Nardos -
Bruce Fischl -
Doug Greve -
Greg Harris -
Hans Johnson -
Nick Schmansky