Freesurfer

freesurfer@nmr.mgh.harvard.edu

21 participants
30070 discussions

Re: [Freesurfer] samseg longitudinal processing - process killed
by Cerri, Stefano 05 Sep '22

05 Sep '22

Hi yvg, It is most likely a memory error problem. You will need to allocate more RAM to your calls. You probably need at least 32GB or RAM for such big images (might be even more). You can process the FLAIR images with mri_robust_template since they are slighly smaller than your T1 images. Stefano Dear Samseg developers, I have very high resolution T1 (512 x 512 x 240; voxel size: 0.469 x 0.469 x 1) and FLAIR images ( 384 x 384 x 128; voxel size: 0.495 x 0.495 x 1) and I would like to process them with the longitudinal processing of samseg but I have encountered different issues: 1.- mri_robust_template with T1 images: the process is killed after the following steps: Registration::findSaturation - computing centroids - computing initial transform -- using translation info - Get Gaussian Pyramid Limits ( min size: 16 max size: -1 ) - Build Gaussian Pyramid ( Limits min steps: 0 max steps: 4 ) - Build Gaussian Pyramid ( Limits min steps: 0 max steps: 4 ) ./samseg_mutua.sh: line 20: 1511261 Killed mri_robust_template ... 2.- mri_robust_template with FLAIR images can be processed without errors, but when I execute run_samseg_long to the FLAIR images, the process is killed without further explanations after the following command: "Transforming points[0.49479165 0.49479164 0.49479165]". Could you please please help me? Best, yvg

1 0

samseg longitudinal processing - process killed
by Yolanda Vives-Gilabert 05 Sep '22

05 Sep '22

External Email - Use Caution Dear Samseg developers, I have very high resolution T1 (512 x 512 x 240; voxel size: 0.469 x 0.469 x 1) and FLAIR images ( 384 x 384 x 128; voxel size: 0.495 x 0.495 x 1) and I would like to process them with the longitudinal processing of samseg but I have encountered different issues: 1.- mri_robust_template with T1 images: the process is killed after the following steps: Registration::findSaturation - computing centroids - computing initial transform -- using translation info - Get Gaussian Pyramid Limits ( min size: 16 max size: -1 ) - Build Gaussian Pyramid ( Limits min steps: 0 max steps: 4 ) - Build Gaussian Pyramid ( Limits min steps: 0 max steps: 4 ) ./samseg_mutua.sh: line 20: 1511261 Killed mri_robust_template ... 2.- mri_robust_template with FLAIR images can be processed without errors, but when I execute run_samseg_long to the FLAIR images, the process is killed without further explanations after the following command: "Transforming points[0.49479165 0.49479164 0.49479165]". Could you please please help me? Best, yvg

1 0

Re: [Freesurfer] Samseg dice is low
by Cerri, Stefano 04 Sep '22

04 Sep '22

Hi Dilek, 0.2 Dice seems extremely low. Also, in my experience, a threshold of 0.05 on T1w-FLAIR scans will show up many FP lesions. We usually reported a Dice in the range of 0.50-0.65 on various datasets (MS and WMH, default threshold, T1w-FLAIR), so I'm surprised by a such low value. Can you share with us your run_samseg command? Stefano Date: Sun, 4 Sep 2022 15:53:17 +0300 From: Dilek ?zkan <dozkan521(a)gmail.com> Subject: To: Freesurfer support list <freesurfer(a)nmr.mgh.harvard.edu> Message-ID: <CAPQSyvLqkvjk7JFxFoAG8RcX7ff40TS70vzvF6uWWHwC6CgqEg(a)mail.gmail.com> Content-Type: text/plain; charset="utf-8" External Email - Use Caution Hello all, I use Freesurfer-Samseg for MS lesion segmentation(cross-sectional). I use the T1 and Flair for this. (just these are inputs) I registered *flair and t1 to DIR *for every patient. (My manual lesion masks are on DIR imaging.) I controlled the registrations, these are true. I created a binary lesion mask using *threshold 0.05. *(mri_binarize --i /data/projects/MS_MRI/posteriors/Lesions.mgz --o /data/projects/MS_MR/binarize/lesion_binary.mgz --min 0.05) I checked the dice score with ground truth values with python. *threshold *0.05= *dice* 0,2 of the dataset Do you have any recommendations to increase the dice? # cvs_version 7.2.0 # sysname Linux # machine x86_64 Best regards

1 0

Samseg dice is low
by Dilek Özkan 04 Sep '22

04 Sep '22

External Email - Use Caution Hello all, I use Freesurfer-Samseg for MS lesion segmentation(cross-sectional). I use the T1 and Flair for this. (just these are inputs) I registered *flair and t1 to DIR *for every patient. (My manual lesion masks are on DIR imaging.) I controlled the registrations, these are true. I created a binary lesion mask using *threshold 0.05. *(mri_binarize --i /data/projects/MS_MRI/posteriors/Lesions.mgz --o /data/projects/MS_MR/binarize/lesion_binary.mgz --min 0.05) I checked the dice score with ground truth values with python. *threshold *0.05= *dice* 0,2 of the dataset Do you have any recommendations to increase the dice? # cvs_version 7.2.0 # sysname Linux # machine x86_64 Best regards

1 0

Re: [Freesurfer] Illegal Instruction (mri_segment_hypothalamic_subunits) - VM 7.2.0
by fsbuild 03 Sep '22

03 Sep '22

External Email - Use Caution Hello Randy, The mri_segment_hypothalamic_subunits command is a python script that uses tensorflow which in turn is coded to use AVX instructions.  If the VM host processor does not support AVX instructions, then you will get an illegal instruction error. You can check to see if AVX instructions are supported if this commands shows "avx" or "avx2" in the output (see screenshot for where that is the case). $ lscpu | grep avxIf AVX instructions are supported, then error can also happen if the command is incomplete. - R.

2 1

Re: [Freesurfer] problem with autorecon2-cp
by Mostafa Fakhri 03 Sep '22

03 Sep '22

External Email - Use Caution Hi Doug, Thank You for your response I have a new problem I used -autorecon1 and autorecon2-cp: recon-all -sd /share/users/mfakhri/freeSurfer-analys/subjects -subject fastWM -i /share/users/mfakhri/freeSurfer-analys/100610_7T/T1w.nii.gz -autorecon1 -autorecon2-cp But process was stopped after half an hour with this error: $ cat /share/users/mfakhri/freeSurfer-analys/subjects/fastWM/mri/error.log mghRead(/share/users/mfakhri/freeSurfer-analys/subjects/fastWM/mri/norm.mgz, -1): could not open file Mostafa

2 1

Request for 7 zip password VM_67
by Cristian Neira 02 Sep '22

02 Sep '22

External Email - Use Caution Hello, I just need a passcode to archive the zip folder for file FS_7_2_0_ubuntu_18_04_06.ova so that I can install a FreeSurfer virtual disk. Thanks, Cristian

1 0

FreeView feature request
by Aaron Tanenbaum 02 Sep '22

02 Sep '22

External Email - Use Caution I am looking set up freesurfer to do two things. If they are not currently possible then I would like to submit two feature requests for freeview. Here they are When making voxel edits, it would be useful to see all voxels the mouse is pointing to. Currently the mouse looks like a pen pointing to one voxel. If my brush size is 5, it would be nice if all 25 voxels are highlighted. It would also be convenient to have the voxels that have been edited in a single brush stroke highlighted.. Frequently I will be making an edit to an image while another image is overlaid on top of it. So I can not see my edits. forcing me to toggle between images. Thank you

1 0

Postdoc: Brain networks, development, and mental illness risk
by Signe Bray 02 Sep '22

02 Sep '22

External Email - Use Caution Drs. Emma Towlson and Signe Bray at the University of Calgary are seeking applications for a postdoctoral scholar who will use c*omputational modeling of brain network alterations to study relationships to mental illness risk*. The project will involve working with neuroimaging data to build whole-brain simulations and examine network characteristics. Depending on interest, the postdoc may play a role in study design, data collection and processing, though the focus is modeling and data analysis. This is a 2-year position with the possibility of extension. Applicants should have interest in computational approaches to brain and mental health and have a background in network neuroscience, psychology, neurosciences, physics, biomedical engineering, or a related discipline. Experience with neuroimaging and strong data analytic skills are required. Calgary is a city of ~1.3M located a short-drive from Canada’s Rocky Mountains, offering outstanding outdoor recreation opportunities and more days of sunshine per year than any other major Canadian city. The postdoc will integrate with more than 300 Brain and Mental Health researchers at UCalgary and benefit from a new Computational Neuroscience Platform ( https://secure-web.cisco.com/17Kog-EAduJ6JkIBsDLKbEmlZaM8f3jyTiFiopeWtCLjVZ…) Interested candidates can find more details here https://secure-web.cisco.com/1UNSbOBLYUqexjCjYi1cQrH3hT9oGInSvtQKwe-O89g9Fp… and submit a letter of interest and CV to slbray(a)ucalgary.ca and emma.towlson(a)ucalgary.ca or reach out with any questions.

1 0

TRACULA - Ball-and-stick model fit running time
by Liliana Wu 02 Sep '22

02 Sep '22

External Email - Use Caution Hi, I have successfully run the preprocessing steps for my DWI data, and I was wondering how long does bedpostx usually take to process DWI? I am running on a high-powered computer (cluster) and a multi-shell data 3 b-values of (b= 1000s/mm2, 34 directions; b= 1600s/mm2, 45 directions; and b= 2600s/mm2, 66 directions). Thank you, Liliana

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