Freesurfer March 2023

freesurfer@nmr.mgh.harvard.edu

104 participants
154 discussions

Is it possible to add additional T1 scans to improve surfaces/segmentation
by Michael Mc Laughlin 01 Apr '23

01 Apr '23

External Email - Use Caution I am working with standard neuro radiological MRI scans. These typically have something like this:: a) T1 headscout volume: 1.6mmx1.6mmx1.625mm 160x160x128 b) T1 se tra 0.72x0.72x5.2 320x320x30 c) T2 tse tra scan 0.72x0.72x5.2 320x320x30 d) T2 FLAIR (dark fluid) scan 0.58x0.58x5.2 460x460x30 c) T1 tse sag 0.72x0.72x5.2 320x320x30 c) T2 tse sag 0.72x0.72x5.2 320x320x30 I have found that the best results are obtained if I do the following: i) Start with the T1 headscout volume: 1.6mmx1.6mmx1.625mm This produces poor pial results but much better than using the T1s with a small number of slices. ii) Add in the flair data: Use -FLAIR and -autorecon3 with a FLAIR MRI scan This dramatically improves the pial surfaces and they look almost normal iii) Add some T2 data. This give more improvement of the pial surfaces See attached image for how the cortex looks after these stages My questions are: *1) How can I add in the data from the higher resolution T1 scans? (with only 30 slices)* *2) If I add a second T2 scan with -T2 -T2pial -autorecon3 will the results improve even more?* Regards, Michael

3 4

White matter exclusions occurring in v7.3.2 but not v5.3
by Keefe, Sarah 31 Mar '23

31 Mar '23

External Email - Use Caution Hi FreeSurfer Team, Our group is working on reprocessing scans that we originally processed (performed quality control and edited where necessary) with FreeSurfer 5.3 with the HCP patch. We are aiming to reprocess everything with FreeSurfer version 7.3.2. So far when checking and editing FreeSurfer 7 outputs we're regularly seeing areas of brain that should be included within the white matter surface but are not, particularly in the accumbens and orbitofrontal areas. The areas are of lower intensity within the scans, but FreeSurfer 5.3 is able to handle these areas accurately - we don't see these issues in our runs on the same scan files done in FreeSurfer 5.3. I've attached a screenshot of the Freesurfer output of each version side by side on one of the slices where the exclusion is happening. We're seeing this on maybe 1 in 10 Freesurfer 7 scans. All did fine in 5.3. We're finding that the problem is not limited to certain scanners - one of these examples is from a Biograph PETMR scan and the other is from a Prisma. Is there a recommended flag or flags we could use with Freesurfer 7 recon-all that might get us closer to the way 5.3 handles this? The Freesurfer 5.3 parameters we originally used were: -all -qcache -hippo-subfields -nocubic -mprage The Freesurfer 7.3.2 parameters we are using now are: -all -qcache I have the recon-all outputs of both versions and the original scan available and can send them via FTP if that would be helpful. Other notes: Running with the -mprage flag added to the Freesurfer 7.3.2 recon-all launch was not different from the FS 7.3.2 recon-all launch without the flag but I can send that output along as well if it would be useful. Also, these examples are done in 7.3.2 but we've also run into this issue with 7.1.1. Really appreciate any assistance you can provide, as trying to figure this out has been delaying our processing a bit. Thanks so much, Sarah ________________________________ 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.

3 7

Synthseg segmentation of skull etc
by Fraser Aitken 31 Mar '23

31 Mar '23

External Email - Use Caution Dear Freesurfer community I have some low field T1 and T2 MRI images that I would like to use to create a head model for EEG source localization. I have been using SynthSR to improve the images and Synthseg for the basic segmentation etc and wondered if anyone had used synthseg to extract skull and scalp as well as WM/GM/CSF etc and how this might be done ? (or if this was possible) best wishes Fraser

2 2

automatic screenshots to all axial slices
by Gonzalo Rojas Costa 31 Mar '23

31 Mar '23

External Email - Use Caution Hi: I need to take screenshots of all axial slices of a volumetric image that I am viewing using freeview (flair with a segmentation). How can I do it automatically? Sincerely, Gonzalo Rojas Costa

2 1

multiple axial slice screenshots
by Gonzalo Rojas Costa 31 Mar '23

31 Mar '23

External Email - Use Caution Hi: I need to take screenshots to all axial slices of a volumetric image that I am viewing using freeview (flair with a segmentation). Sincerely, Gonzalo Rojas Costa

1 0

moving results from fsaverage space to subject space
by Jennifer Kilmer 31 Mar '23

31 Mar '23

External Email - Use Caution Hello— I am having difficulties figuring out how to move PET images from the fsaverage space back to the subject space. Using the PETSurfer wiki instructions for surface-based analysis, we have sampled each subject’s surface onto the surface of fsaverage using mri_vol2surf, which was then smoothed using mris_fwhm, and finally the results were brought back from surface space to volume (using mri_surf2vol), still in the fsaverage space. We are doing an analysis in SPM comparing each individual subject to controls and would like to move these results to the subject space, because remaining in the fsaverage space has given us results that do not match up very well and make the analysis difficult. We have attempted using mri_vol2vol but the resulting gray matter ribbon did not align with our subject T1. Is it necessary to combine the talairach lta/m3z register files and/or inverse them to achieve this? and if so, how? or if there is a simpler way to move the results from fsaverage space back into subject space, please let me know. Thank you in advance for your help! Jennifer

2 1

Re: [Freesurfer] HPC Usage - Moving On from CentOS7
by fsbuild 31 Mar '23

31 Mar '23

External Email - Use Caution We don’t use Alma OS which I believe is somewhere downstream from the official RedHat CentOS8 Stream distribution, i.e., the libs may not be the same.  I also don’t see that RedHat CentOS8* releases include CentOS7 versions of libraries. - R. On Mar 30, 2023, at 12:12, Horn, Mitchell Jacob <MJHORN(a)mgh.harvard.edu> wrote:As an update I’ve tested another run (bert, 7.2.0, no flags, recon-all) on an OS-8 system but set LD_PRELOAD to the 7 system’s libm.so: export LD_PRELOAD=/alma/cos7lib/libm.sorecon-all …. And that produced the equivalent results of all my CentOS7 runs: <image002.png> I believe this is very important information to the general community for stability with the imminent and inevitable switch to OS-8 HPC platforms. Best,Mitch_______________________________________________Freesurfer mailing listFreesurfer@nmr.mgh.harvard.eduhttps://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

2 2

Re: [Freesurfer] Samseg lesion mask holes
by Cerri, Stefano,PHD 31 Mar '23

31 Mar '23

Hi Evie, --save-probabilities and --save-posteriors will save two different things for lesions. The former is an initialization of lesions based on intensity and location constraints, and the latter is a probability map of seg.mgz after we also introduce lesion shape constraints (Cf Fig.4 of https://www.sciencedirect.com/science/article/pii/S1053811920309563) The output of --save-probabilities should give you a very noisy lesion probability map with lots of false positives (at least this is my experience). In your case, it seems that too many *true* lesions are removed when the shape model is introduced. Would you be able to provide the command you're using to segment your T2-FLAIR scans? Which --lesion-mask-pattern values are you using for this input combination? Stefano Evie Neylon Tue, 28 Mar 2023 07:15:17 -0700 External Email - Use Caution Dear experts, I am running SAMSEG to segment MS lesions from T2 FLAIR scans. I wanted to create a lesion mask in order to obtain lesion volumes and lesion counts, so I used mri_binarize on seg.mgz to obtain all voxels with the value 99. However, upon visualising this mask I see that there are holes in the mask in regions that are quite clearly lesions based on the FLAIR image. I see in seg.mgz that these voxels are instead assigned incorrectly to a different tissue, as the lateral ventricle, for example. I wondered if this misassignation of tissue was due to the chosen threshold (0.3) so I added the --save-probabilties flag to look at the probability maps too, and I see that actually the probability map thresholded at 0.3 does not contain holes. The probability map actually seems more accurate in general and picks up on more of the hyperintensities. This is important for my desired outputs (i.e. accurate lesion counts/volumes). So I wondered if it was possible to explain a little why these holes might be occurring in the final segmentation but not the probability map? Would it be incorrect to derive the lesion mask from the probability map instead? Apologies I cannot attach the images of the data. TIA, Evie _______________________________________________ Freesurfer mailing list Freesurfer(a)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 Mass General Brigham Compliance HelpLine at https://www.massgeneralbrigham.org/complianceline <https://www.massgeneralbrigham.org/complianceline> . Please note that this e-mail is not secure (encrypted). If you do not wish to continue communication over unencrypted e-mail, please notify the sender of this message immediately. Continuing to send or respond to e-mail after receiving this message means you understand and accept this risk and wish to continue to communicate over unencrypted e-mail.

2 2

Regressing out unconditioned stimulus
by Gillian Grennan 30 Mar '23

30 Mar '23

External Email - Use Caution Hi Freesurfer Developers, I am analyzing a fear conditioning imaging study and am attempting to regress out the shock/unconditioned stimulus by adjusting the modulation weights in my par file. I set the stimuli presented with the shock (CS+) with a modulation weight of -1 and set all other unshocked/unpaired stimuli (CS- and fixation) with a weight of 1. Similarly, we have a few CS+ trials where the participants are expecting a shock but didn't receive one -- I have been setting these with a modulation weight of 1 similar to the CS- and fixation stimuli. I attached a copy of our .par file. Could you let me know if this makes sense for our analyses? Best, Gillian Grennan

2 1

kvlThresholdImage: command not found
by Song, Xinrui 30 Mar '23

30 Mar '23

External Email - Use Caution Hello Dear Developers, I am trying to segment ex vivo hemisphere brains following instructions here "https://secure-web.cisco.com/1aMllqwj8kw45W6STdIjlcZduOuYMIQpzI8UIeO-4ITFYp…". I was just proceeding to the Segment White Matter section and found that none of the kvl commands can be found in my installation. They are also not found in the freesurfer/bin folder. I searched the archive and found that there was one case that's very similar to mine, but it occurred back in 2012 and on a Mac distribution. (https://secure-web.cisco.com/1l-Cmn5p7-rI6gQsdH_JGo18SYVHYKwuPuWwgNbpZRVRkl…) Freesurfer version: freesurfer_ubuntu20-7-dev_amd64.deb Platform: WSL-Ubuntu on Windows.

2 1

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Freesurfer March 2023