Freesurfer

freesurfer@nmr.mgh.harvard.edu

19 participants
30088 discussions

GPU/supercomputer adaptation of freesurfer?
by John Absher 14 Dec '18

14 Dec '18

External Email - Use Caution Hi, I'm planning a freesurfer analysis of a large MRI dataset, and want to use the "380 GPU nodes" (and other cores/nodes) on the Palmetto Cluster (https://www.palmetto.clemson.edu/palmetto/userguide_palmetto_overview.html) to speed up the process. Since I am not a programmer, I'm hoping someone can give me a quick tutorial: a) Is this going to speed up recon-all and the data analysis? b) How much programming/expertise is required to enable freesurfer to take advantage of a supercomputer's resources? c) Has anyone done this already? d) The Palmetto Cluster is more or less limited to command-line. As long as I visualize the data on another system, I assume this will not be a problem, right? Thanks, John R. Absher, MD jabsher(a)ghs.org<mailto:jabsher@ghs.org> GHS Neuroscience Associates University of South Carolina School of Medicine Greenville 864-350-6655 (mobile)

4 4

Longitudinal stream : LME and limits of model
by Matthieu Vanhoutte 14 Dec '18

14 Dec '18

Dear Freesurfer’s experts, I would have some questions regarding the LME model to be used in longitudinal stream: 1) Which are the ratio limits or % of missing timepoints accepted ? (according time, I have less and less subjects time points) 2) Is it possible to include patients that would miss the first timepoint but got the others ? 3) Considering a group in longitudinal study, which is the number of subjects minimal of this group accepted for LME modeling ? 4) Finally, concerning quality control and among a big number of total time points, which essential controls are necessary ? (Control of norm.mgz of the base, alignment of longitudinal timepoints on base,… ?) Best regards, Matthieu

2 13

Longitudinal processing - LAS and RAS orientation baseline vs follow-up
by Elena Pozzi 14 Dec '18

14 Dec '18

External Email - Use Caution Dear FreeSurfer Experts, I have some baseline and follow-up MRI scans that I would like to process through the longitudinal stream. Some of the baseline images were originally in Analyze format (LAS orientation) and were converted into nifti (keeping the LAS orientation) and then processed through the recon-all stream (and edited). A few other baseline images were instead in nifti format (RAS orientation). The follow-up images were all in nifti format (RAS orientation) and were as well processed through the recon-all stream and edited. For all the images that were originally in Analyze format at baseline and converted in nifti, the within subject template (base) failed, i.e., the surfaces clearly do not properly follow the WM and GM boundaries. Instead, for the baseline images that were in nifti format the template looks fine. Given that the problem arose only with the Analyze images, could the different orientation (LAS and RAS) explain the issue? If so, is it possible to flip the orientation after the preprocessing, keeping the edits that were made on the pial of the baseline images? Apologies if any of these questions was already asked but I couldn’t find an answer. Thank you very much, Elena

2 1

Longitudinal Stream - LONG step
by Figueiro Longo, Maria Gabriela 14 Dec '18

14 Dec '18

Hello FreeSurfer Developers, I'm attempting to run the longitudinal stream, but in the [LONG] step of one time-point of on of my subjects, I have the following error: MRInormInit(): INFO: Modifying talairach volume c_(r,a,s) based on average_305 MRInormalize(): MRIsplineNormalize(): npeaks = 11 Starting OpenSpline(): npoints = 11 MRInormFindControlPoints: could not find enough control points building Voronoi diagram... performing soap bubble smoothing, sigma = 8... Iterating 2 times --------------------------------- 3d normalization pass 1 of 2 No such file or directory MRInormFindControlPoints failed No such file or directory The segmentation of the this time-point for this subject ([CROSS] step) looks good and I couldn't find any error. I didn't find any similar error in the email list. Does anyone have any thoughts on how to trouble-shoot this one? I attached the recon-all.log. FS version: freesurfer-linux-centos7_x86_64-dev-20181130-24625d8 Thanks, Gabriela

3 3

Longitudinal processing and paralllel flag
by Erik O'Hanlon 14 Dec '18

14 Dec '18

External Email - Use Caution Hi FS experts, I'm running a longitudinal process with three time points per subject and was wondering if you can advise on the typical time to process? Can the -parallel flag be added to the base and long steps? I'm running this on a cluster that has 16 core nodes with 32gigs of ram so it might give me a good tme saving if these steps can be parallelized. Thanks and regards Erik Postdoctoral Research Fellow Dept. Of Psychiatry Royal College of Surgeons in Ireland ERC Building Beaumont Hospital Dublin & Dept. Of Psychiatry Trinity College Institute of Neuroscience The Llyod Institute Trinity College Dublin D2 Erik O'Hanlon Postdoctoral researcher [cid:rcsi-crest-signature_f581c185-1d03-45c8-a786-23d8ece3d391.png] RCSI Psychiatry Royal College of Surgeons in Ireland Beaumont Road, Beaumont D9 Ireland T: 8093740 E: erikohanlon(a)rcsi.ie W: www.rcsi.com<http://www.rcsi.com/> Transforming Healthcare Education, Research and Service: RCSI Strategic Plan 2018-2022<http://www.rcsi.ie/strategy2018> [cid:AS_BRONZE_IIDD_65c26b0a-ea45-42c8-956e-5d84e884e4e6.png]

2 1

longitudinal analyses after lesion segmentation
by kathleen Gerbino 14 Dec '18

14 Dec '18

External Email - Use Caution Hi FreeSurfer experts: I hope you are well. I need to segment quite a few longitudinal T1 scans from patients with multiple sclerosis. A colleague has applied an automated lesion segmentation algorithm to segment and fill the lesions. During this process, the follow-up scans were registered to the baseline scan. My question is whether I could still apply the Freesurfer longitudinal pipeline to these images or whether your had any recommendations on the appropriate steps to take if this is not appropriate. Kind regards, Kathleen

2 2

Does data orientation (Neurological Vs. Radiological) produces different results
by Meena M. Makary 13 Dec '18

13 Dec '18

External Email - Use Caution Hi; I am wondering if the orientation of the input data to recon-all (i.e. Neurological Vs. Radiological) produces different results in terms of laterality (right vs. left)? Thanks; --Meena — *Meena M. Makary, Ph.D.* Assistant Professor | Cairo University <http://bmes.cufe.edu.eg/> Postdoctoral Fellow | Yale University School of Medicine <http://people.yale.edu/search/meena_beshara.profile> Postdoctoral Associate | The John B. Pierce Laboratory <http://jbpierce.org/postdoctoral-associates/> Postdoc Committee Chair | The John B. Pierce Laboratory <http://jbpierce.org/postdoctoral-associates/> Career Development and Mentoring Manager Elect | OHBM Student and Postdoc SIG <https://www.humanbrainmapping.org/i4a/pages/index.cfm?pageid=3449> 290 Congress Avenue, New Haven, CT 06519 +1 (475) 731-5592 <+1%20475-731-5592>

2 1

Re: [Freesurfer] How to use FIR to get HRF shape
by Lauri Tuominen 13 Dec '18

13 Dec '18

External Email - Use Caution I want to test if there is an increase in the HRF delay in a certain condition. I would like the time resolution to be higher so that the time to peak is measured as accurately as possible. I thought this would be the idea of not locking the events with the TR. Am I completely lost here? Cheers! Lauri Hi Lauri, what is non-ideal about it? doug On 12/06/2018 05:30 PM, Lauri Tuominen wrote: > External Email - Use Caution > > Hi there, > > I would like to know if a condition leads to a delayed HRF peaking. > > To device an experiment to answer this, I collected finger tapping pilot data with 80 trials of 3s of tapping, and jittered ITI between 5 to 25 seconds, TR=1.6. > > The trials are not TR locked because I thought I?d sample different portions of the HRF during each trial. > > I run the mkanalysis-sess with -fir 4.8 25.6 flag and all looks to be fine. > > My problem is that the resulting ces.nii.gz has only 16 volumes, which correspond to the TR, I assume (because 25.6/1.6=16). So using this, my time resolution for peak detection would be 1.6, which is not what I would like it to be. This is obviously less than ideal for answering my research question. > > How could I get the FIR fit to each timepoint so I could reconstruct the HRF myself? Or would there be some other workaround / analysis? Or do I need to collect the data in some other way, i.e somehow sliding the TR-event locking? > > Additional question: what is in cesmag.nii.gz, cesmagpct.nii.gz, iminsig.nii.gz and minsig.nii.gz? > > Thanks so much again for the help > > Lauri Tuominen >

2 1

Re: [Freesurfer] Recon command with T1 and T2
by Maksimovskiy, Arkadiy 13 Dec '18

13 Dec '18

Got it. Thank you!

1 0

Re: [Freesurfer] Recon command with T1 and T2
by Maksimovskiy, Arkadiy 13 Dec '18

13 Dec '18

Hi Doug, I have a follow up question, if you don’t mind clarifying. My goal is to optimize the reconstruction and increase its accuracy. Would you mind confirming if the flag selection is appropriate for the following scenarios: Multiple T1 scans and one T2 scan: Use the 3T flag and the -T2pial flag (is combining them allowed?). One T1 scan and one T2 scan: Use only the -T2pial flag . I would love to hear if these are correct. Thank you for your help, arkadiy

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