Hi Doug,
I asked a question a couple days ago for how to convert .label files into .nii files in freesurfer. At the end of this email, you can see the original email from me and your response.
I tried to run the function mri_label2vol by myself. However, it seems not working quite well.
In order to run mri_label2vol, it seems that I have to have a registration.dat file. I search online for how to generate it and get an answer from maillinglist posted by you several years ago. That is, I need to run the function of tkregister2. Then I tried to find out how to run tkregister2 and I found out an example suggested by you, although I don't know how it works because there is no information online for this function.
Your example is like this:
*tkregister2 --mov rawavg.mgz --targ aseg.mgz --regheader --reg rawreg.dat --noedit*
I ran it and it successfully generated a rawreg.dat file, but I don't know if it is a correct file that I should use. Below is the results that I got from the command window:
*tkregister_tcl /opt/freesurfer/lib/tcl/tkregister2.tcl target volume aseg.mgzmovable volume rawavg.mgzreg file rawreg.datLoadVol 0ZeroCRAS 0$Id: tkregister2.c,v 1.117.2.1 2010/08/02 17:42:03 greve Exp $Diagnostic Level -1INFO: loading target aseg.mgz Ttarg: ---------------------1.000 0.000 0.000 128.000; 0.000 0.000 1.000 -128.000; 0.000 -1.000 0.000 128.000; 0.000 0.000 0.000 1.000;INFO: loading movable rawavg.mgzTmov: -------------------- -0.900 0.000 0.000 100.799; 0.000 0.000 0.898 -115.000; 0.000 -0.898 0.000 115.000; 0.000 0.000 0.000 1.000;mkheaderreg = 1, float2int = 0Computing reg from header (and possibly input matrix) ---- Input registration matrix (computed) ---------1.000 -0.012 -0.025 -0.000;-0.025 0.069 0.997 0.000; 0.010 -0.998 0.070 -0.000; 0.000 0.000 0.000 1.000;--------------------------------------- ---- Input registration matrix ---------1.000 -0.012 -0.025 -0.000;-0.025 0.069 0.997 0.000; 0.010 -0.998 0.070 -0.000; 0.000 0.000 0.000 1.000;Determinant -1subject = subject-unknown RegMat ----------------------------1.000 -0.012 -0.025 -0.000;-0.025 0.069 0.997 0.000; 0.010 -0.998 0.070 -0.000; 0.000 0.000 0.000 1.000;*
Then I tried to run the function of mri_label2vol. I specifically referred to the example posted online in freesurfer website. It seems to me that this is a popular function that a lot of people have used. My command line is as below:
*mri_label2vol --label lh.words10_10.label --temp T1.mgz --reg rawreg.dat --fillthresh .5 --o lh.words10_10.nii*
However, my command line didn't run successfully to generate the nii file that I want. The error message is "segmentation flaut". Here I attached all the results from the command window after I run this function.
*Number of labels: 1lh.words10_10.labelAnnot File: (null)Template Volume: T1.mgzOutut Volume: lh.words10_10.niiRegistration File: rawreg.datFill Threshold: 0.5Label Vox Vol: 1ProjType: (null) ProjTypeId: 0ProjStart: 0ProjStop: 0ProjDelta: 0.1Subject: (null)Hemi: (null)UseNewASeg2Vol: 1DoLabelStatVol 0$Id: mri_label2vol.c,v 1.32 2009/10/22 23:35:53 greve Exp $ Template RAS-to-Vox: ---------1.000 0.000 0.000 128.000;-0.000 -0.000 -1.000 128.000;-0.000 1.000 -0.000 128.000; 0.000 0.000 0.000 1.000;Template Voxel Volume: 1nHits Thresh: 0.5 Loading registration from rawreg.datRegMat: ---------1.000 -0.012 -0.025 -0.000;-0.025 0.069 0.997 0.000; 0.010 -0.998 0.070 -0.000; 0.000 0.000 0.000 1.000;Label RAS-to-Vox: -------- 1.000 0.012 0.025 128.000;-0.010 0.998 -0.070 128.000;-0.025 0.069 0.997 128.000; 0.000 0.000 0.000 1.000;Allocating Hit Volume (16777216) voxelsnlabels = 1Loading lh.words10_10.label Thesholding hit volume.Segmentation fault*
I was surprised by this result, because it tells me that "the template voxel volume is 1". How strange! And finally, no nii file was generated.
Can you help check whether I did something wrong? Attachment I put the files that might be used for running these two functions (I didn't put T1.mgz and rawavg.mgz in this email since it makes the file bigger than 500K, but if you need it please let me know and I will send you the link from my dropbox). I might need to specify more parameters in the functions, which I didn't how to either. If you can supervise me for how to write the functions correctly, that would be really appreciated.
And after I generate nii file, how can I visualize it and see how it matches the ROI from the .label file? I tried tksurfer "subject name" lh inflated, but it doesn't work. If I can get some instructions for this command, that would be also helpful.
Thanks.
Best, Jingjing
Date: Tue, 29 Oct 2013 10:21:59 -0400 From: Douglas N Greve greve@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] how to transfer .label file to .nii file To: freesurfer@nmr.mgh.harvard.edu Message-ID: 526FC487.6020601@nmr.mgh.harvard.edu Content-Type: text/plain; charset=windows-1252; format=flowed
You can use mri_label2vol to create a nii mask from the label. To map the label to the DTI space, register your DTI (b=0) to the anatomical using bbregsiter. You can then use mri_label2vol to transfer the label.
doug
On 10/28/2013 06:14 AM, Zhao Jingjing wrote:
Hi,
Our lab have managed to generate ROIs in T1 images using Freesurfer, which led to .label files. We succesfully measured the cortical thickness in these ROIs for each subject in Freesurfer. Now we are aiming to track white matter fibers from these ROIs in a tractography software Trackvis. However, .lable files seem not be readable in Trackvis. Thus, we are wondering if .label files can be tranferred to .nii files easily in Freesurfer so that we can use the converted .nii ROI files to track fibers in Trackvis. Another related question is how can we coregister the ROIs drawn in T1 images with DTI images and make sure that those ROIs converted from T1 images are the still in the same regions in the DTI images?
Thanks. Jingjing Zhao
Postdoc scientist
Laboratoire de Sciences Cognitives et Psycholinguistique
Departement d?Etudes Cognitives
Ecole Normale Superieure
29 rue d?Ulm, 75230, Paris Cedex 05, France
Tel: +33 1 44 32 26 57
Fax: +33 1 44 32 26 30