Freesurfer

freesurfer@nmr.mgh.harvard.edu

19 participants
30088 discussions

local_gi problem
by krista kelly 12 Feb '14

12 Feb '14

Hello, I am trying to do local gyrification with freesurfer and am having trouble. I have the image processing toolbox and have set the path. I ran recon-all -s BV21 -localGI and it exited without errors. Yet, when I try to view it in tksurfer using tksurfer BV21 lh inflated -overlay lh.pial_lgi, I get this error: cannot find rh.pial_lgi (even though it is in the surfs folder). I can open the inflated brain in tksurfer and overlay lh.pial_lgi but there seems to be a problem where not all of the brain is colored in (please see attached pictures). I'm wondering if someone can help me with this problem, it's happened on more than one participant. When checking through the terminal, I found this at the beginning of the processing: INFO: FreeSurfer build stamps do not match Subject Stamp: freesurfer-Linux-centos4-stable-pub-v5.3.0 Current Stamp: freesurfer-Darwin-lion-stable-pub-v5.3.0 I initially did recon-all -all using a linux on a virtual machine, however there wasn't enough working memory to do all of my analyses using linux, so I transferred all of my data to a mac and did the quality checks, extracted roi data (aparcstats2table) and ran glm analyses (qdec) from there. Do you think that the problem lies in the fact that recon-all was originally ran in Linux? I'd rather not do the local gyrification in Linux as I do not have matlab on that machine. Thanks! Krista

2 1

Postdoc - ICM Paris - Morphometry using 7T MRI
by Olivier Colliot 12 Feb '14

12 Feb '14

Dear colleagues, Our team is looking for a postdoctoral fellow. Please find below the informations about the position. Many thanks Olivier Colliot _*Research topic: Computational anatomy of the hippocampus using very high resolution MRI at 7 Tesla: diffeomorphic registration and shape analysis *_The hippocampus is an anatomical structure of the brain which plays critical roles in different brain disorders including Alzheimer's disease and epilepsy. The hippocampus has the particularity of presenting a highly complex tridimensional organization with a large number of substructures, fields and layers. Current approaches for shape analysis of the hippocampus only consider its external boundary and do not model its complex internal architecture. Indeed, using conventional MRI at 1.5T or 3T, only the external boundaries of the hippocampus can be visualized and its internal subregions remain inaccessible. The advent of ultra-high field MRI (7T and higher) has opened new perspectives by providing increased signal-to-noise ratio, increased tissular contrast and increased spatial resolution, which allows visualizing in vivo tiny anatomical structures (~100-200 microns). New computational anatomy approaches are needed to handle these complex anatomical structures. The aim of this postdoctoral project is to develop new approaches for diffeomorphic registration and shape analysis of the internal structures of the hippocampus using 7T MRI data with very high spatial resolution. The developments will be based on the framework of large diffeomorphic deformations and matching of currents. New developments based on varifolds and functional currents will be performed in order to handle the complex shapes of the internal structures. The work will be done in collaboration with our partners at ENS de Cachan (Alain Trouvé) and Université Paris-Descartes (Joan Glaunès). The developed approaches will be applied to detect subtle alterations in patients with Alzheimer's disease and epilepsy in which 7T MRI has been acquired. These clinical applications will be done in collaboration with our partners at Neurospin, Pitié-Salpêtrière hospital, and University of Minnesota. _*Profile of the candidate *_The ideal candidate should have previous experience in the design of new methods for morphometry, shape analysis or registration of anatomical medical images. However, we will also consider outstanding scientific profiles in the fields of image processing/computer vision, without previous experience of the specific topics mentioned above. _*About the ARAMIS team*_ ARAMIS is a joint research team between CNRS, INRIA, Inserm and UPMC within the Brain and Spine Institute (ICM). The ICM (http://icm-institute.org/menu/foundation/mission?lang=en) is a recently created neuroscience center based in the Pitié-Salpêtrière hospital in Paris, which is the largest adult hospital in Europe and has a long tradition of neuroscience and neurology. The ICM brings together 500 researchers covering the whole field of neuroscience. ARAMIS is the methodological research team of the ICM. It is a multidisciplinary group (electrical engineering, computer science, neurology) with about 30 members. The team aims at designing new mathematical and computational approaches for the analysis of images and signals of the human brain. The key domains of expertise of the team are: i) automatic segmentation of brain structures; ii) shape analysis and longitudinal modeling; iii) machine learning; iv) analysis of brain connectivity. The team has developed advanced methods for the analysis of anatomical and functional neuroimaging data including the fully automatic approach SACHA to segment the hippocampus and the amygdala, pattern classification algorithms, the Deformetrica software for shape analysis, and graph-theory approaches to functional connectivity. The team also has strong expertise in the management of multi-center neuroimaging studies, including multi-site MRI harmonization, quality control and analysis of massive datasets. The team has close and enduring collaborations with different clinical teams of the ICM and the Pitié-Salpêtrière hospital to apply advanced image analysis approaches to neurological disorders, including Alzheimer's disease, epilepsy, fronto-temporal dementia, Parkinson's disease, Gilles de la Tourette syndrome. _*Contact information*_ Olivier Colliot - olivier.colliot(a)upmc.fr - http://cogimage.dsi.cnrs.fr/perso/colliot/ _*More information:*_ https://www.inria.fr/en/institute/recruitment/offers/post-doctoral-research…

1 0

Running a template through recon-all
by Danielle Sliva 12 Feb '14

12 Feb '14

Dear FreeSurfer experts, I am trying to map volumetric ROIs obtained from VBM to surfaces in 5-6 year old children. I saw in a couple previous posts that it is recommended to run the template brain through recon-all, map the ROIs to the template surface, and then propagate the ROIs to all subjects using either mri_surf2surf, or (after turning the ROIs into labels) using mri_label2label. The problem I have is that we used a custom age-appropriate template that will not pass through recon-all. I have tried using -notalairach & -notal-check. Here is the output from the recon-all.log: register_mri: find_optimal_transform find_optimal_transform: nsamples 3481, passno 0, spacing 8 GCAhistoScaleImageIntensities: could not find wm peak resetting wm mean[0]: 117 --> 126 resetting gm mean[0]: 74 --> 74 input volume #1 is the most T1-like using real data threshold=24.0 skull bounding box = (128, 99, 138) --> (128, 154, 185) using (128, 117, 162) as brain centroid... mean wm in atlas = 126, using box (128,110,156) --> (127, 123,167) to find MRI wm WARNING: gca.c::GCAhistoScaleImageIntensities: h_mri->nbins=255, mri_peak=-1 before smoothing, mri peak at 0 WARNING2: gca.c::GCAhistoScaleImageIntensities: h_mri->nbins=255, mri_peak=-1 after smoothing, mri peak at 0, scaling input intensities by inf Linux rc-twice 3.2.21-mosix #1 SMP Mon Jul 2 08:55:48 EDT 2012 x86_64 x86_64 x86_64 GNU/Linux recon-all -s TOM_5yr_2 exited with ERRORS at Mon Feb 10 19:11:23 EST 2014 I am using version 5.3. Can you suggest any other options to get this template through recon-all? Thanks in advance, Danielle

1 0

How to avoid tumour from being segmented out
by Melanie Morrison 12 Feb '14

12 Feb '14

Hi there, I am creating surfaces from my volume T1-weight scans so that I can run them in AFNI's SUMA program. Problem is that my tumour has been segmented out because its intensity and the intensity of its surround area is much lower that what is being segmented. How can I avoid this? Or edit this? I need my functional MRI data to be displayed around the tumour region which is now unfortunately cut out. (HELPP!!!) Thanks, Melanie Morrison

2 1

freesurfer course
by Jayachandra Raghava 12 Feb '14

12 Feb '14

Hello, I am interested in attending a freesurfer course at Martinos Center for Biomedical Imaging in Boston<http://maps.google.com/maps?q=%22114%20Sixteenth%20Street,%20Charlestown,%2…> But, it says that the following lectures are only during spring and fall. 1)Functional analysis (during the Spring course) 2)Diffusion processing (during the Fall course) I would like to attend the course which covers the above topics. Can you please give me the link for the details of the workshop which includes diffusion and functional course. Regards Jay

1 0

distribute dots on the surface
by peng 12 Feb '14

12 Feb '14

Dear Freesurfers, I have run recon-all and got a set of surfaces for the subjects. I wish to distribute N dots about evenly on the surface to cover most gyri and sulci , but I don't how to implement that. One way is to manually put them in the GUI of freeview, but it is not so convenient if N is relative large (say, 200). Can anyone give me a hint which function shall I use? best Peng

2 1

Re: [Freesurfer] Can -min_border_white and -min_gray_at_white_border be edited for single subject in a large group of subjects?
by Bruce Fischl 12 Feb '14

12 Feb '14

Hi Erik yes, those numbers are always estimated from the data unless you specify otherwise. And please remember to cc the list so that others can answer/see the answers in case they are interested in the same type of issue. cheers Bruce On Wed, 12 Feb 2014, Erik Lindberg wrote: > OK Thanks > - I suspected that. What made me a bit confused what that I went in and had > a look at the recon-all log file. And numbers from this file was different > for different subjects at the same batch. > > > > > > > two subjects > setting MIN_GRAY_AT_WHITE_BORDER to 67.9 (was 70) > setting MAX_BORDER_WHITE to 107.3 (was 105) > setting MIN_BORDER_WHITE to 79.0 (was 85) > setting MAX_CSF to 56.7 (was 40) > setting MAX_GRAY to 94.7 (was 95) > setting MAX_GRAY_AT_CSF_BORDER to 73.4 (was 75) > setting MIN_GRAY_AT_CSF_BORDER to 45.6 (was 40) > > > setting MIN_GRAY_AT_WHITE_BORDER to 71.5 (was 70) > setting MAX_BORDER_WHITE to 108.1 (was 105) > setting MIN_BORDER_WHITE to 82.0 (was 85) > setting MAX_CSF to 61.0 (was 40) > setting MAX_GRAY to 95.9 (was 95) > setting MAX_GRAY_AT_CSF_BORDER to 76.7 (was 75) > setting MIN_GRAY_AT_CSF_BORDER to 50.5 (was 40) > > > On Wed, Feb 12, 2014 at 2:31 PM, Bruce Fischl <fischl(a)nmr.mgh.harvard.edu> > wrote: > Hi Erik > > is there a reason not to do it for the whole group? I guess if > the one subject you uploaded had significant motion (which > caused the blurryness) then it might be the case that it needs > separate parameters. In general though it's good to hold this > kind of thing constant as it will definitely bias you to having > greater thickness. Try it on some of the other subjects and see > if the surfaces are at least as accurate as without, and if so > do it uniformly > > cheers > Bruce > > On Wed, 12 Feb 2014, Erik Lindberg wrote: > > Dear Freesufers, > I am trying to understand what happens if I go in > and edit the numbers for > -min_border_white -min_gray_at_white_border in a > single subjects in a > large group. > > Generally we aim to set everything identical when > running a group of > subjects in FS. I went in and had a look at the > recon-all log file and I > can see that the numbers for -min_border_white > -min_gray_at_white_border different for different > subjects. > > Thus - what will happened if I go and edit the > -min_border_white > -min_gray_at_white_border numbers for a single > subjects in a large bach > .... will I still be able compare cortical thickness > with subjects that I > did not edit? > > Or is it necessary to keep the number identical for > all subjects when > editing the min_border_white > -min_gray_at_white_border > > > > > 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 Partners Compliance > HelpLine at > http://www.partners.org/complianceline . If the e-mail was sent to you > in error > but does not contain patient information, please contact the sender > and properly > dispose of the e-mail. > > > >

1 0

Can -min_border_white and -min_gray_at_white_border be edited for single subject in a large group of subjects?
by Erik Lindberg 12 Feb '14

12 Feb '14

Dear Freesufers, I am trying to understand what happens if I go in and edit the numbers for -min_border_white -min_gray_at_white_border in a single subjects in a large group. Generally we aim to set everything identical when running a group of subjects in FS. I went in and had a look at the recon-all log file and I can see that the numbers for -min_border_white -min_gray_at_white_border different for different subjects. Thus - what will happened if I go and edit the -min_border_white -min_gray_at_white_border numbers for a single subjects in a large bach .... will I still be able compare cortical thickness with subjects that I did not edit? Or is it necessary to keep the number identical for all subjects when editing the min_border_white -min_gray_at_white_border

2 1

Segmentation using Free Surfer.
by Saurabh Thakur 12 Feb '14

12 Feb '14

Hello Free surfer Expert, I have started working on Free surfer. Now i m facing two issues such as : 1. I m trying to use contour ... but it is working as per guidance in website given. https://surfer.nmr.mgh.harvard.edu/fswiki/FreeviewGuide/FreeviewTools/Voxel… Under heading of Contour Draw....can any tell me pre requisite to draw contour. I have loaded T1 image and then made new volume....after tat ctrl+alt+ left click....not working...? Can any one help me with contour function, how to use in details. 2. I wanted to do segmentation of image into WM,GM and CSF. Please suggest what the best way/ Algorithm to so get MRI image segmentation using Freesurfer. Thanks in advance. cheers Saurabh Thakur,

2 1

Hemispherectomy Surgery
by Kristian Eschenburg 11 Feb '14

11 Feb '14

Hi I have a dataset of children who have had hemispherectomy surgery to treat epilepsy. I want to be able to run recon-all on these children. I have masked out the remaining tissue from the excised side so that only tissue from the healthy hemisphere remains, and aligned this single hemisphere to a template. To mimic a whole brain, I have flipped the single hemisphere over the X-axis and added it to the other half. Thus, I have the original hemisphere, and the "mirrored" hemisphere in one scan. This is my input to recon-all. However, as the scans were acquired some time after surgery, the remaining hemisphere was not perfectly aligned at the midline and expanded / grew into to the empty cranial space, and there are gaps and overlap between the mirrored side and the original side. Is there a way to run recon-all on the single hemisphere alone, so that I do not have to mirror the brain? Perhaps by supplying a mask, to apply to the FreeSurfer atlases? Thanks Kristian

2 2

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