Freesurfer

freesurfer@nmr.mgh.harvard.edu

15 participants
30064 discussions

recon-all error
by Cara Bohon 23 Aug '17

23 Aug '17

I have a couple of subjects "exited with ERRORS" after running almost 48 hours (when my others all took about 13-15 hours on the same server). Both exited during Fix Topology lh. The only time "error" appears is in this line: IFLAG= -1 LINE SEARCH FAILED. SEE DOCUMENTATION OF ROUTINE MCSRCH ERROR RETURN OF LINE SEARCH: INFO= 4 POSSIBLE CAUSES: FUNCTION OR GRADIENT ARE INCORRECT OR INCORRECT TOLERANCESoutof QuasiNewtonEMA: 011: -log(p) = -0.0 tol 0.000010 - which prior messages to this list suggest is not meaningful or problematic. I am brand new to Freesurfer, so I'm not quite sure where to start in troubleshooting. The orig.mgz file looks fine to me, but maybe I'm missing something. Any other tips for things to check? I'm attaching a recon-all.log file in case that is helpful. I am running Freesurfer 5.3.0 on Linux. Thanks for any tips on addressing the error! -Cara

2 1

Longitudinal stream for cross-sectional analysis?
by K 23 Aug '17

23 Aug '17

Good morning Martin et al., We are analyzing cortex and subcortical volumes for 2 groups with (mostly) 2 scans per participant. We want to report longitudinal findings as well as cross-sectional group differences at each of the 2 time points. For the time point 1 and time point 2 analyses, does it make sense to use the longitudinal stream-processed data, or is it better to use the cross-sectional stream for each time point independently? For example: Comparing the brain stem volumes for individual participants using longitudinal vs cross-sectional data for time point 1, we notice some differences (cross-sectional volumes tend to be 95-98% as large, varying), and group differences are more pronounced when using the longitudinal data (i.e., some structures did not show significant group differences in volumes using cross-sectional data, but did show significant differences using longitudinal). I know you have written about reduced variability and better precision with the longitudinal stream, but we weren't sure if this also applied to using it only at one time point to report cross-sectional findings. Appreciate your guidance! K Additional keywords: longitudinally, cross-sectionally, single time point, benefits, advantages, drawbacks, cortical thickness, surface area, volumetric

2 1

Longitudinal freesurfer
by neuroimage analyst 23 Aug '17

23 Aug '17

Hi, We are currently doing longitudinal FreeSurfer analysis on subjects some of whom are scanned over a period of 6 years. There were 2 groups. One was the control group and the other was mTBI group. The hypothesis was the control group would have a fairly stable cortical and sub-cortical volume (<1% change over time) while mTBI group would have a decreasing volume. Both the groups are relatively young (most participants between 22-30 years old). Unfortunately, we only have 7 controls that were scanned for over 3 years. What we have observed is there is no pattern in change in volume neither for controls nor for mTBI; with volume increasing from baseline to year1 and then decreasing and then for some increasing again. The delineation looks ok to us. There are a lot of subjects ~120 and hence manual delineation on every subject is not practical. The question is what is the expected measurement variance due to the longitudinal algorithm for healthy controls over time in both cortical and subcortical volumes? Since the effect that we were expecting to see is around 4% in mTBI participants and the delineation too looks okay then what is the best procedure to correct these? OR should we use cross-sectional data for our analysis? We have used longitudinal freesurfer v6 on a centos. I look forward to your thoughts. Thanks Regards Virendra

2 1

Longitudinal spatiotemporal LME vs. V-LME
by Ann-Kathrin Rogge 23 Aug '17

23 Aug '17

Dear freesurfer experts, I have a mixed 2x2 design with group (approx. 18 subjects per group) and timepoint (t0, t1) as factors. I analyzed the thickness data with the longitudinal stream and the spatiotemporal LME as a random intercept model, age as covariate. However, the model does not converge perfectly, although I varied the size of the homogenous covariance regions between 1-2 (parameter k, Bernal-Rusiel et al., 2013). In addition, when I apply a voxelwise-LME, more clusters survive the FDR correction. It seems that the V-LME is more sensitive than the SP-LME in my case (clusters at an uncorrected threshold are more or less the same). I would love to understand why. Could anyone explain this to me? And would it be statistically valid to apply the V-LME to my data? Best, Ann-Kathrin -- *Ann-Kathrin Rogge, Dipl.-Psych.* www.bpn.uni-hamburg.de

2 1

longitudinal generic-time and spc
by Kasper Jessen 23 Aug '17

23 Aug '17

Dear Freesurfer, The following is a follow-up, on a previous discussion (http://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg51582.html) but i am still left with some uncertain and unanswered questions. My design: Patients and Controls, MRI scanned at baseline and after 6 weeks (only two time points). The average follow-up time is about 7 weeks when using the exact time. 1) I understand that the if you pass the --generic-time in the long_stats_slopes command, the 'time' column should not be in the qdec table as it assumes the values are 1,2,3,4 (subjects are ordered in TP1, TP2). However, if there is an exact time, then a 'time column' should be in the qdec table and the --generic-time should not be passed in the long_stats_slopes command. Is this correct? 2) If the exact time is not present, how is the symmetrized precent change calculated? Since it is calculated spc = 100 * rate / avg, and rate is calculated: rate = ( thick2 - thick1 ) / (time2 - time1) then there must be a time right? Thanks in advance. I hope some of you have find the time to help :-) I am a little stuck with these details. Best regards, Kasper Sent from [ProtonMail](https://protonmail.com), Swiss-based encrypted email.

2 1

Error after manual edits in freesurfer 6
by Chethana 23 Aug '17

23 Aug '17

To, The Freesurfer experts, I am analysing a set of MRI images in freesurfer 6. In the step 1 of freesurfer data processing the dura and skull are stripped and during the process of inspection with manual edits of the aux volume if I save the progress and reopen the same image for review they appear very brightly intense and unable to differentiate grey from white matter. Attaching two pictures for the same the first picture is without saving The second brightly intense picture attached is when the edit is opened for review after saving the volume where the grey white matter differentiation is difficult to see Can you kindly suggest how do I rectify this error or any other solution for the same? Thanks and regards Chethana

2 1

Hard failure via XNAT pipeline (Freesurfer 6)
by Rockers, Elijah D. 23 Aug '17

23 Aug '17

We encountered a hard failure, below is the tail of the log. We are currently running Freesurfer 6 again locally to see if it succeeds on a desktop machine (this is the first time encountering a hard failure on the XNAT server). Any suggestions? -merging segment 180 into 178 160 defects to be corrected 0 vertices coincident reading input surface /data/build/nnac/20170818_114535/alamode_MRI08162017/20170818114538/alamode_MRI08162017/surf/lh.qsphere.nofix... reading brain volume from brain... reading wm segmentation from wm... Computing Initial Surface Statistics -face loglikelihood: -8.7982 (-4.3991) -vertex loglikelihood: -6.2982 (-3.1491) -normal dot loglikelihood: -3.5611 (-3.5611) -quad curv loglikelihood: -6.1931 (-3.0965) Total Loglikelihood : -24.8505 CORRECTING DEFECT 0 (vertices=40706, convex hull=11265, v0=89) XL defect detected... Segmentation fault (core dumped) Linux hhvrixnatp01 2.6.32-431.el6.x86_64 #1 SMP Sun Nov 10 22:19:54 EST 2013 x86_64 x86_64 x86_64 GNU/Linux recon-all -s alamode_MRI08162017 exited with ERRORS at Mon Aug 21 19:27:09 CDT 2017 To report a problem, see http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting Houston Methodist. Leading Medicine. Houston Methodist is ranked by U.S. News & World Report as the No. 1 hospital in Texas for patient care. Houston Methodist is nationally ranked in 8 specialties and is designated as a Magnet hospital for excellence in nursing. Visit us at houstonmethodist.org. Follow us at twitter.com/MethodistHosp and facebook.com/HoustonMethodist. ***CONFIDENTIALITY NOTICE*** This e-mail is the property of Houston Methodist and/or its relevant affiliates and may contain restricted and privileged material for the sole use of the intended recipient(s). Any review, use, distribution or disclosure by others is strictly prohibited. If you are not the intended recipient (or authorized to receive for the recipient), please contact the sender and delete all copies of the message. Thank you.

2 1

Re: [Freesurfer] hippocampal and amygdalar subregional segmentation
by Iglesias Gonzalez, Eugenio 23 Aug '17

23 Aug '17

Hi Moji: 1. Absolutely. You can use standard FreeSurfer tools for extracting masks from a segmentation, resample them to whatever space, etc. 2. A lot of people have been making this question, so we’re now writing a bunch of merged segmentations to the mri directory of the subjects for convenience. If none of these groupings satisfies you, it is always possible to use standard FreeSurfer tools to do this. 3. Yes, the new atlas includes the nuclei of the amygdala (Saygin, Kliemann et al, NeuroImage, 2017). Features 2 and 3 will be available in the dev version in the next few days; stay tuned! Cheers, /Eugenio -- Juan Eugenio Iglesias ERC Senior Research Fellow Translational Imaging Group University College London http://www.jeiglesias.com http://cmictig.cs.ucl.ac.uk/ From: <freesurfer-bounces(a)nmr.mgh.harvard.edu> on behalf of "Aghajani, Moji" <m.aghajani(a)vumc.nl> Reply-To: Freesurfer support list <freesurfer(a)nmr.mgh.harvard.edu> Date: Wednesday, 23 August 2017 at 12:14 To: "freesurfer(a)nmr.mgh.harvard.edu" <freesurfer(a)nmr.mgh.harvard.edu> Subject: [Freesurfer] hippocampal and amygdalar subregional segmentation Dear Freesurfer Developers, I have some questions regarding hippocampal and amygdalar subregional segmentation in the latest Freesurfer version: 1) Can one use the segmentation outputs as ROI masks, register those to native space functional image, and extract functional time courses in order to perform seed-based subregional functional connectivity analyses? 2) How sensible do you deem it to assess the functional connectivity profiles of all the subregions? Would you argue against grouping some of the subregions into a larger subregion and assess connectional profiles. For instance combining CA 1 to 4 into a CA subgroup? Or the Subiculum and Pre- Parasubiculum into one subgroup? I am asking this because most literature on hippocampal subregional connections focuses on the major subregions (CA, Dentate Gyrus, Subiculum), with very limited knowledge on connectivity profiles of smaller individual subregions. 3) Is it already possible to segment the amygdala subregions as well with this new pipeline? On the Freesurfer website the following is stated: “In addition to the hippocampal subregions, it also simultaneously segments the nuclei of the amygdala”. However, the page also states that this functionality will soon be available in the dev version, so I am not sure whether the amygdala subregional segmentation pipeline is already in place? If not, is there a preliminary release date known already? Thanks for your time and effort!

1 0

hippocampal and amygdalar subregional segmentation
by Aghajani, Moji 23 Aug '17

23 Aug '17

Dear Freesurfer Developers, I have some questions regarding hippocampal and amygdalar subregional segmentation in the latest Freesurfer version: 1) Can one use the segmentation outputs as ROI masks, register those to native space functional image, and extract functional time courses in order to perform seed-based subregional functional connectivity analyses? 2) How sensible do you deem it to assess the functional connectivity profiles of all the subregions? Would you argue against grouping some of the subregions into a larger subregion and assess connectional profiles. For instance combining CA 1 to 4 into a CA subgroup? Or the Subiculum and Pre- Parasubiculum into one subgroup? I am asking this because most literature on hippocampal subregional connections focuses on the major subregions (CA, Dentate Gyrus, Subiculum), with very limited knowledge on connectivity profiles of smaller individual subregions. 3) Is it already possible to segment the amygdala subregions as well with this new pipeline? On the Freesurfer website the following is stated: “In addition to the hippocampal subregions, it also simultaneously segments the nuclei of the amygdala”. However, the page also states that this functionality will soon be available in the dev version, so I am not sure whether the amygdala subregional segmentation pipeline is already in place? If not, is there a preliminary release date known already? Thanks for your time and effort!

1 0

Error after manual edits in freesurfer 6
by Chethana 23 Aug '17

23 Aug '17

To, The Freesurfer experts I am analysing a set of MRI images in freesurfer 6. In the step 1 of freesurfer data processing the dura and skull are stripped and during the process of inspection with manual edits of the aux volume if I save the progress and reopen the same image for review they appear very brightly intense and unable to differentiate grey from white matter. Attaching two pictures for the same the first picture is without saving The second brightly intense picture attached is when the edit is opened for review after saving the volume where the grey white matter differentiation is difficult to see Can you kindly suggest how do I rectify this error or any other solution for the same? Thanks and regards Chethana

1 0

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