Freesurfer May 2011

freesurfer@nmr.mgh.harvard.edu

120 participants
155 discussions

RAS to MNI coordinates of a label
by Yolanda Vives 06 May '11

06 May '11

Hi all, I am interested in obtaining the MNI or talairach coordinates of a label. The procedure that I have performed is the following: First of all I have performed a QDEC analysis. From the results, I have drawn a label of a significant area of the brain, that corresponds to the cingulate. Afterwards, I have used the matlab function read_label to obtain the RAS coordinates of this label. To transform these coordinates into mni coordinates, I have read that if v = [R A S 1]' and X matrix is the matrix from talairach.xfm, then vMNI = X*v Taking this into account, I have used the X matrix from the talairach.xfm of "fsaverage" (where the results of qdec are mapped). This matrix is 1.0 0.0 0.0 0.0 0.0 1.0 0.0 0.0 0.0 0.0 1.0 0.0 so vMNI is the same as v. When I introduce v (or vMNI) coordinates into a MNI atlas, I don't obtain the cingulate. Could you please help me with this procedure? Thank you in advance, Yolanda

4 8

Smaller Virtual Machine
by Fidel 05 May '11

05 May '11

Hi all, I have been trying the virtual disk image, and it works well. Nevertheless, I think that more than 7GB is a lot of space if somebody wants to use it as a Virtual Machine in a cloud environment, in which you are charged for used storage and maybe for used bandwidth in the data communication. Is anybody thinking about creating a lighter version of this virtual disk image? Thanks and best regards, Fidel.

2 5

How to proceed after manual edit of segmentation
by Tetiana Dadakova 05 May '11

05 May '11

Dear FreeSurfer list, I am making some manual correction to subcortical structures using Edit Segmentation Tool. And my question is: how should I proceed after saving aseg.mgz? Should I re-run autorecon2 with some options? Thank you, Tanja.

3 6

Re: [Freesurfer] Interpreting results from Qdec
by Gallen, Courtney (NIH/NIDA) [F] 05 May '11

05 May '11

Hi Doug Have a follow up question regarding significance levels in Qdec related to extracting means from ROIs, etc. I'm posting to the entire mailing list in case anyone else has ideas. Basically, I extracted mean volumes (from my set of participants) of ROIs for which Qdec reports an interaction between two variables. I want to look at these mean volumes in something like SPSS to see what is driving the interaction/generate figures, etc. I'm having an issue with one such ROI, in that when I extract the mean volumes and look for this interaction in SPSS, it is not significant (p ~ 0.5, so really not significant at all). Is there any way that an ROI could be significant in Qdec (p of 0.05, Monte Carlo sim), but not in SPSS when I extract means for each participant? Let me know if this isn't clear. Any thoughts would be appreciated! Courtney Courtney Gallen Post-baccalaureate IRTA Neuroimaging Research Branch National Institute on Drug Abuse (IRP) 251 Bayview Blvd Suite 200 Baltimore, MD 21224 Tel: (443) 740-2631 -----Original Message----- From: Douglas Greve [mailto:greve@nmr.mgh.harvard.edu] Sent: Tuesday, April 26, 2011 1:06 PM To: Gallen, Courtney (NIH/NIDA) [F] Subject: Re: [Freesurfer] Interpreting results from Qdec Those are the values outside (first column) and inside (2nd col) the ROI. You can add "--id 1" to only report the values inside the ROI. doug On 4/26/11 1:02 PM, Gallen, Courtney (NIH/NIDA) [F] wrote: > It worked! > Now, last question. The txt file has two separate values (?averages). What do they each represent? > > -----Original Message----- > From: Douglas Greve [mailto:greve@nmr.mgh.harvard.edu] > Sent: Tuesday, April 26, 2011 12:53 PM > To: Gallen, Courtney (NIH/NIDA) [F] > Subject: Re: [Freesurfer] Interpreting results from Qdec > > try "--slabel fsaverage lh ROI-drawn-on-QDEC-brain.label" > > > On 4/26/11 12:52 PM, Gallen, Courtney (NIH/NIDA) [F] wrote: >> Actually, I hope I'm wrong about what I thought below. If I specify individual subject's label and run: >> >> mri_segstats --in /prot/Mprot/CTQ/freesurferDone/8579/surf/lh.volume.fwhm10.fsaverage.mgh --slabel 8579 lh /prot/Mprot/CTQ/freesurferDone/8579/label/lh.SupTemp_CTQxBDNFctrls.label --avgwf lhSupTempctrls.txt >> >> I still get the same error about dimension mismatch >> >> >> -----Original Message----- >> From: Gallen, Courtney (NIH/NIDA) [F] >> Sent: Tuesday, April 26, 2011 12:43 PM >> To: 'Douglas Greve' >> Subject: RE: [Freesurfer] Interpreting results from Qdec >> >> Ah I think I see the problem now. >> >> The label I originally specified is the general label created from the ROI drawing in Qdec. Instead, I should use the individual subject's label (from 'Map Label to Subjects') instead, right? >> >> Thank you! >> >> -----Original Message----- >> From: Douglas Greve [mailto:greve@nmr.mgh.harvard.edu] >> Sent: Tuesday, April 26, 2011 12:38 PM >> To: Gallen, Courtney (NIH/NIDA) [F] >> Subject: Re: [Freesurfer] Interpreting results from Qdec >> >> You have to use the subject that matches the input, fsaverage in this >> case. How did you define your label? If you defined it on subject 123, >> you can transfer it to fsaverage using mri_label2label. >> doug >> >> On 4/26/11 12:28 PM, Gallen, Courtney (NIH/NIDA) [F] wrote: >>> Not a problem at all! >>> >>> If I use --in, I get an error saying that there is a dimension mismatch between input volume and seg. >>> >>> Any other thoughts? Thanks again! >>> >>> -----Original Message----- >>> From: Douglas Greve [mailto:greve@nmr.mgh.harvard.edu] >>> Sent: Tuesday, April 26, 2011 11:40 AM >>> To: Gallen, Courtney (NIH/NIDA) [F] >>> Subject: Re: [Freesurfer] Interpreting results from Qdec >>> >>> >>> Hi Courtney, sorry for the delay. If more than a few days go by without >>> a response, feel free to repost -- I promise not to get annoyed! >>> Sometimes, things get pushed up and out of my mail browser window, and I >>> often don't get back to them. Sorry! >>> >>> To answer your question, try using --in where you have --seg. The >>> "segmentation" is actually already specified because you've specified a >>> label. >>> >>> doug >>> >>> On 4/26/11 11:03 AM, Gallen, Courtney (NIH/NIDA) [F] wrote: >>>> Hi Doug, >>>> Sorry to keep bugging you but any thoughts about my response to your e-mail from 2 weeks ago? I've given the exact command line I'm using with mri_segstats. >>>> Thanks again for all your help >>>> Courtney >>>> >>>> -----Original Message----- >>>> From: Gallen, Courtney (NIH/NIDA) [F] >>>> Sent: Tuesday, April 12, 2011 10:59 AM >>>> To: 'Douglas N Greve' >>>> Subject: RE: [Freesurfer] Interpreting results from Qdec >>>> >>>> That may be the problem. ${sub} is an actual person's directory >>>> >>>> E.g., sub# 123--command line would be: >>>> mri_segstats --seg /prot/Mprot/CTQ/freesurferDone/123/surf/lh.volume.fwhm10.fsaverage.mgh --slabel 123 lh \ >>>> lh.SupTemp_CTQxBDNFctrls.label --avgwf lhSupTempctrls.txt >>>> >>>> >>>> >>>> -----Original Message----- >>>> From: Douglas N Greve [mailto:greve@nmr.mgh.harvard.edu] >>>> Sent: Tuesday, April 12, 2011 10:55 AM >>>> To: Gallen, Courtney (NIH/NIDA) [F] >>>> Subject: Re: [Freesurfer] Interpreting results from Qdec >>>> >>>> What is your full command-line? Without variables, just something that >>>> you could put directly into the terminal. Make sure that you are >>>> specifying fsaverage as your subject (can't tell below because your use >>>> ${sub}). >>>> >>>> doug >>>> >>>> >>>> Gallen, Courtney (NIH/NIDA) [F] wrote: >>>>> Sorry, but I can't seem to figure out how to specify input data. I tried '--in' but get an error that the input vol and seg label are not the same dimensions. I've been looking at the help options for mri_segstats (http://surfer.nmr.mgh.harvard.edu/fswiki/mri_segstats) but don't see it >>>>> >>>>> -----Original Message----- >>>>> From: Douglas N Greve [mailto:greve@nmr.mgh.harvard.edu] >>>>> Sent: Friday, April 08, 2011 10:38 AM >>>>> To: Gallen, Courtney (NIH/NIDA) [F] >>>>> Subject: Re: [Freesurfer] Interpreting results from Qdec >>>>> >>>>> The lh.volume.fwhm10.fsaverage.mgh should be your input. It is not a >>>>> segmentation. You've already specified the seg with the the --slabel >>>>> command. >>>>> doug >>>>> >>>>> >>>>> Gallen, Courtney (NIH/NIDA) [F] wrote: >>>>> >>>>>> One last question (hopefully). I'm specifying the input volume with --seg, but am getting an error ('cannot do frame average without input volume'). Am I missing something obvious? >>>>>> >>>>>> The entire command looks something like this: >>>>>> mri_segstats --seg /prot/Mprot/CTQ/freesurferDone/${sub}/surf/lh.volume.fwhm10.fsaverage.mgh --slabel ${sub} lh \ >>>>>> lh.SupTemp_CTQxBDNFctrls.label --avgwf lhSupTempctrls.txt >>>>>> >>>>>> Thanks again >>>>>> -----Original Message----- >>>>>> From: Douglas N Greve [mailto:greve@nmr.mgh.harvard.edu] >>>>>> Sent: Tuesday, April 05, 2011 4:21 PM >>>>>> To: Gallen, Courtney (NIH/NIDA) [F] >>>>>> Cc: Nick Schmansky; Freesurfer Mailing List >>>>>> Subject: Re: [Freesurfer] Interpreting results from Qdec >>>>>> >>>>>> yes, something like ?h.volume.fwhm10.fsaverage.mgh >>>>>> doug >>>>>> >>>>>> Gallen, Courtney (NIH/NIDA) [F] wrote: >>>>>> >>>>>> >>>>>>> Hi Doug >>>>>>> >>>>>>> Thanks for helping me with this! >>>>>>> >>>>>>> First time around I did everything as is outlined in 'Define a Region of Interest' here: http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/QdecGroupAnalysis. >>>>>>> >>>>>>> I'll try your suggestion now. The input data would be something like ?h.volume.fwhm10.fsaverage.mgh, correct? >>>>>>> >>>>>>> Thanks again >>>>>>> Courtney >>>>>>> >>>>>>> -----Original Message----- >>>>>>> From: Douglas N Greve [mailto:greve@nmr.mgh.harvard.edu] >>>>>>> Sent: Thursday, March 31, 2011 5:44 PM >>>>>>> To: Gallen, Courtney (NIH/NIDA) [F] >>>>>>> Cc: Nick Schmansky; Freesurfer Mailing List >>>>>>> Subject: Re: [Freesurfer] Interpreting results from Qdec >>>>>>> >>>>>>> Hi Courtney, did you transfer the label to each individual subject using >>>>>>> mri_label2label? A better way to replicate your qdec results is to run >>>>>>> mri_segstats on the input data to qdec (y.mgh usually). Specify the >>>>>>> label with "--slabel subject hemi yourlabel". Also specify "--avgwf >>>>>>> subjectdata.txt". This will create this text file with a list of the >>>>>>> input data for each subject averaged over your label. >>>>>>> >>>>>>> doug >>>>>>> >>>>>>> Gallen, Courtney (NIH/NIDA) [F] wrote: >>>>>>> >>>>>>> >>>>>>> >>>>>>>> Hi Nick >>>>>>>> >>>>>>>> I'm revisiting this issue and have a few more questions. First, I have a significant interaction between two variables on cortical volume. I drew a label on this ROI and ran mris_anatomical_stats on the label (thank you for that suggestion). >>>>>>>> >>>>>>>> For stats in the output table file, is the column 'GrayVol' the same thing as the cortical volume I looked at in Qdec? I'm asking because when I export GrayVol values into SPSS, there is no longer a significant interaction between my variables (and if this ROI is significant after multiple comparisons across the brain, I'd expect the mean volume from the ROI to be very significant in SPSS). >>>>>>>> >>>>>>>> Please let me know if my questions aren't clear. Thanks in advance >>>>>>>> Courtney >>>>>>>> >>>>>>>> >>>>>>>> Courtney Gallen >>>>>>>> Post-baccalaureate IRTA >>>>>>>> Neuroimaging Research Branch >>>>>>>> National Institute on Drug Abuse (IRP) >>>>>>>> 251 Bayview Blvd >>>>>>>> Suite 200 >>>>>>>> Baltimore, MD 21224 >>>>>>>> Tel: (443) 740-2631 >>>>>>>> >>>>>>>> >>>>>>>> -----Original Message----- >>>>>>>> From: Nick Schmansky [mailto:nicks@nmr.mgh.harvard.edu] >>>>>>>> Sent: Thursday, January 13, 2011 3:08 PM >>>>>>>> To: Gallen, Courtney (NIH/NIDA) [F] >>>>>>>> Cc: Freesurfer Mailing List >>>>>>>> Subject: Re: [Freesurfer] Interpreting results from Qdec >>>>>>>> >>>>>>>> Courtney, >>>>>>>> >>>>>>>> if i understand your question, i think one way to do that is to use qdec >>>>>>>> to draw a label on the blob of interest, and the run 'map label to >>>>>>>> subjects', which creates a label file in each subjects label dir, then >>>>>>>> you can run stats on that with mris_anatomical_stats. >>>>>>>> >>>>>>>> (btw, i'm putting this answer on the list, as others may have better >>>>>>>> ideas.) >>>>>>>> >>>>>>>> n. >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> On Mon, 2011-01-03 at 16:38 -0500, Gallen, Courtney (NIH/NIDA) [F] >>>>>>>> wrote: >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>>> Hi Nick >>>>>>>>> >>>>>>>>> Thanks for the prompt reply. Sorry if my questions seem a bit naïve--this is my first time using Freesurfer. >>>>>>>>> >>>>>>>>> A follow up question to your answer for question 1. I see that the group data will be plotted in Qdec for significant blobs, but say there is a significant interaction between two variables and it's not entirely apparent what is driving this interaction. >>>>>>>>> Is there a way to extract the mean of this blob for each individual (i.e., treat it as an ROI or something similar) to determine what's driving the significance? >>>>>>>>> >>>>>>>>> Thanks in advance! >>>>>>>>> Courtney >>>>>>>>> >>>>>>>>> -----Original Message----- >>>>>>>>> From: Nick Schmansky [mailto:nicks@nmr.mgh.harvard.edu] >>>>>>>>> Sent: Thursday, December 30, 2010 2:44 PM >>>>>>>>> To: Gallen, Courtney (NIH/NIDA) [F] >>>>>>>>> Cc: freesurfer(a)nmr.mgh.harvard.edu >>>>>>>>> Subject: Re: [Freesurfer] Interpreting results from Qdec >>>>>>>>> >>>>>>>>> ans. 1. - the easiest way is to Ctrl- left mouse click on a blob, and a >>>>>>>>> plot of the data at that surface vertex will appear. the group will be >>>>>>>>> apparent from that (say, demented group is thinner than non-demented >>>>>>>>> group). this data is significance data (log p, so '2' is 0.01), so mean >>>>>>>>> and stdev wouldnt apply in that case. the file 'y.mgh' is the raw data >>>>>>>>> of all subjects in the analysis, so you could extract mean and stdev >>>>>>>>> from that. >>>>>>>>> >>>>>>>>> ans. 2. - slide the 'annotation' opacity slider to show the annotation >>>>>>>>> data (cortical parcellation). the ctrl left click will also put the >>>>>>>>> region info for that vertex on the lower left of the display. >>>>>>>>> >>>>>>>>> see also: >>>>>>>>> http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/QdecGroupAnalysis >>>>>>>>> >>>>>>>>> n. >>>>>>>>> >>>>>>>>> On Wed, 2010-12-29 at 10:49 -0500, Gallen, Courtney (NIH/NIDA) [F] >>>>>>>>> wrote: >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>>> Hi, >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> I’m currently running analyses in Qdec and have identified areas that >>>>>>>>>> show significant effects in my contrasts. I have two questions related >>>>>>>>>> to interpreting these results. >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> 1. How can you interpret the effect in each “blob?” (i.e., which >>>>>>>>>> group has a greater cortical surface area, etc.). Right now, all I >>>>>>>>>> know is that there is an effect in certain areas. Can you export this >>>>>>>>>> data to get means and standard errors? >>>>>>>>>> >>>>>>>>>> 2. Is there a way to localize where these effects are? E.g., a >>>>>>>>>> tool that tells you this effect is located in __ area of the brain? >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> Thanks in advance >>>>>>>>>> >>>>>>>>>> Courtney >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> Courtney Gallen >>>>>>>>>> >>>>>>>>>> Post-baccalaureate IRTA >>>>>>>>>> >>>>>>>>>> Neuroimaging Research Branch >>>>>>>>>> >>>>>>>>>> National Institute on Drug Abuse (IRP) >>>>>>>>>> >>>>>>>>>> 251 Bayview Blvd >>>>>>>>>> >>>>>>>>>> Suite 200 >>>>>>>>>> >>>>>>>>>> Baltimore, MD 21224 >>>>>>>>>> >>>>>>>>>> Tel: (443) 740-2631 >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> _______________________________________________ >>>>>>>>>> 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 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. >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>> _______________________________________________ >>>>>>>> Freesurfer mailing list >>>>>>>> Freesurfer(a)nmr.mgh.harvard.edu >>>>>>>> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer >>>>>>>> >>>>>>>> _______________________________________________ >>>>>>>> Freesurfer mailing list >>>>>>>> Freesurfer(a)nmr.mgh.harvard.edu >>>>>>>> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >

5 5

Logic behind mri_vol2surf
by Tuomas Tolvanen 05 May '11

05 May '11

Dear freesurfer experts, I was wondering that what is the logic or technique how mri_vol2surf works? I mean that I have an spmT-image that contains T-values for voxels and I want to visualize these activations on a surface. The number of voxels with value > 3 is ~60. When I check the number of vertices with value > 3, it's ~250 (after mri_vol2surf). So how are the voxels projected to the surface (vertices)? -Tuomas

2 1

nearest mesh vertex
by Gabriele Arnulfo 05 May '11

05 May '11

Dear freesurfer experts, my scene is as follow: i need to find the nearest vertex in the pial surface to a subset of specified control points. I've already checked all the transformation matrices and the points are correctly located either in scanner RAS and in tkrRAS. Now the problem is that using mesh_vertex_nearest freesurfer tool or dsearchn/ DelaunayTri class matlab tools the resulting vertexes are not so correctly located as they visually would be. I'll attach some screenshots from freeview for better understanding. As you guys can easily see in the attached png ,the cursor is located at the "nearest vertex" position while the control point, which is from the left the second red dot, is at least located at the opposite side of the right hemisphere. This doesn't happen for ALL the points but only for some of them like this one. Is there a more reliable way to compute automatically the nearest vertex for a specified control point? Thank in advance, -- Gabriele Arnulfo PhD Student Bioengineering Department of Systemistic, Telematic and Informatic University of Genoa, Italy Viale Causa 13, 16100

2 5

freeview and control points label
by Gabriele Arnulfo 05 May '11

05 May '11

Dear freesurfer experts, I'm currently using freeview to visualize surfaces and volumes. I'm wondering if there's a possibility to assign to each control point a label. My control points are contacts of linear-intra-cerebral electrodes and I would like to assign a name to each control points in order to easily navigate through the brain volume and find the expected contact without inserting the ras coordinate to a cursor position. Is there a smart way to do this? Thank in advance, -- Gabriele Arnulfo PhD Student Bioengineering Department of Systemistic, Telematic and Informatic University of Genoa, Italy Viale Causa 13, 16100

2 2

A function to read mgh files into R
by Heath Pardoe 05 May '11

05 May '11

Hi all, I've written a function that reads *.fsaverage.mgh files into the free statistical software environment "R" (http://www.r-project.org/) and thought it might be useful for other users. Apologies if this has already been done, searching for "R" is kind of non-specific. Any feedback is welcome. Best wishes, Heath Example usage (if you save the function as a text file called load_mgh.R). The output is a list with the thickness values stored in output.name$x > source("load_mgh.R") > lh.thickness <- load.mgh("lh.thickness.fwhm10.fsaverage.mgh") > lh.thickness$x[1:10] [1] 3.0016668 2.9998362 2.4326897 2.2146511 3.3020024 2.2749729 2.6460392 [8] 2.9406335 0.7304581 3.0975394 --------------------------------------------------------------------------- # load mgh file into R load.mgh <- function(input.file) { to.read <- file(input.file,"rb") v <- readBin(to.read,integer(), endian = "big") ndim1 <- readBin(to.read,integer(), endian = "big") ndim2 <- readBin(to.read,integer(), endian = "big") ndim3 <- readBin(to.read,integer(), endian = "big") nframes <- readBin(to.read,integer(), endian = "big") type <- readBin(to.read,integer(), endian = "big") dof <- readBin(to.read,integer(), endian = "big") close(to.read) to.read <- file(input.file,"rb") dump <- readBin(to.read,double(),size = 4,n = 71, endian = "big") x <- readBin(to.read,double(),size = 4, n = ndim1, endian = "big") close(to.read) list(x = x,v = v,ndim1 = ndim1,ndim2 = ndim2,ndim3 = ndim3,nframes = nframes,type = type,dof = dof) } ------------------------------------------------------------------------------ -- Heath R. Pardoe, PhD Brain Research Institute Melbourne Brain Centre 245 Burgundy Street, Heidelberg 3084 Phone: (+61 3) 903 57053 Email: h.pardoe(a)brain.org.au

2 1

unable to use cuda 4.0.13 and cuda 3.2.16 with nvidia 270.41.03, fedora 14
by Knut J Bjuland 04 May '11

04 May '11

It appears that fressurfer 5.0 is compiled against cuda 3.2.x rather than cuda 4. Is there any workaround to force freesurfer 5.0 apart from recompilation for it use cuda 4.0. AS the latest version of the nvidia driver has change cuda version from 3.2 to 4.0. Knut J Cuda 4.0 *[knutjbj@super-knut lib64]$ mri_em_register_cuda Acquiring CUDA device Using default device Segmentation fault (core dumped) Cuda 3.2 [knutjbj@super-knut Downloads]$ mri_em_register_cuda Acquiring CUDA device Using default device CUDA Error in file 'devicemanagement.cu' on line 46 : CUDA driver version is insufficient for CUDA runtime version. Output from deviceQueryDrv CUDA Device Query (Driver API) statically linked version There is 1 device supporting CUDA Device 0: "GeForce GTX 460" CUDA Driver Version: 4.0 CUDA Capability Major/Minor version number: 2.1 Total amount of global memory: 1072889856 bytes Multiprocessors x Cores/MP = Cores: 7 (MP) x 48 (Cores/MP) = 336 (Cores) Total amount of constant memory: 65536 bytes Total amount of shared memory per block: 49152 bytes Total number of registers available per block: 32768 Warp size: 32 Maximum number of threads per block: 1024 Maximum sizes of each dimension of a block: 1024 x 1024 x 64 Maximum sizes of each dimension of a grid: 65535 x 65535 x 65535 Maximum memory pitch: 2147483647 bytes Memory Bus Width: 256-bit Memory Clock rate: 1800.00 Mhz Texture alignment: 512 bytes GPU Clock rate: 1.40 GHz Concurrent copy and execution: Yes # of Asynchronous Copy Engines: 1 Run time limit on kernels: Yes Integrated: No Support host page-locked memory mapping: Yes Concurrent kernel execution: Yes Device has ECC support enabled: No Device is using TCC driver mode: No PASSED Press ENTER to exit... *

2 4

ICV smaller than the brain
by Jeff Sadino 04 May '11

04 May '11

Hello, I asked a question similar to this last year when working with FS 4.3.1, and I'm hoping that there is something new in FS 5 that can help. I am having a problem where a lot of my subjects' brains are larger than the eICV. In FS 4.3.1, it is happening in about 70/140 subjects, and in FS 5, it is happening in about 30/140 subjects. I checked them in tkregister2, and they look ok. I can use the --old-etiv-only command, but then the problem is that my ICV shoots up to 1.7L, and my brain/ICV ratio drops to .65, way to low to be reasonable. What could the reason for this be, and are there any other tricks that I could use? Thank you, Jeff Sadino

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 May 2011