I downloaded two volumes from BrainWeb database. One is treated as ground truth because it has preassigned values(0-11) for each types of tissues and by a simple coding binary brain mask can be achieved. Another volume is the processed one using any brain extraction algorithm. Now, I have two volume one is standard and another processed one. I want to observe the performance of the algorithm that how much accurate which needs to compare both the volumes, pixel by pixel. This comparison will take place properly only when both volumes will generate same number of slides. I am stuck here. Now, ground truth is generating equal number of slides as of the processed volume but these are getting cropped (please see the attached jpg file in previous mail) . Hi Gunjan
you need to give us more information. What are the two volumes? Why is one "ground truth"? What in general are you trying to achieve?
cheers Bruce On Wed, 19 Nov 2014, Gunjan Gautam wrote:
I have attached both the volumes (reference and input) with my mail. Its the last step of my work, I'm stuck at and not finding any solution. I read that FreeSurfer can solve my problem.
I will be obliged if you find some time for this.
Best Regards, Gunjan
On Nov 19, 2014 8:19 PM, "Bruce Fischl" fischl@nmr.mgh.harvard.edu wrote: no, because I don't know what the volume are that you are using On Wed, 19 Nov 2014, Gunjan Gautam wrote:
Ok I try this out. Did you answer my Query 1(related to cropped output of mri_convert command ? I probably did not understand. On Nov 19, 2014 8:10 PM, "Bruce Fischl" <fischl@nmr.mgh.harvard.edu> wrote: Hi Gunjan FreeSurfer has been designed to work as a pipeline. Picking out individual pieces and running them may or may not work, and certainly won't be as reliable. I would advise you to run recon-all -all, then you can use the contents of the subject directory to do whatever analyses you want (e.g. generate an un-bias-corrected, skull stripped volume). The subject id is one that you make up - it will be the name of the directory under which the results are stored. Also specify a SUBJECTS_DIR on the command line, which will be the root directory under which all the individual subject trees live (with -sd <subjects directory path> cheers Bruce On Wed, 19 Nov 2014, Gunjan Gautam wrote: Actually, I have two volumes, belong to the same subject. One volume is the ground truth which will be considered as standard for the second volume which is the processed one or the extracted brain. To perform a perfect comparison between both volumes, these must generate equal number of slices in order to compare the corresponding slide. But here there is problem with my volumes, that is while performing slicing Ground Truth volume is generating 360 slice and processed volume is generating 181 slides. So comparison can't take place in this way. This is the reason I used mri_convert command. If I am not wrong then recon-all command will do bias correction as well, which is not my requirement. What I want is to have the binary brain mask volume and brain volume only. recon-all command is asking for the subject-id and I'm not sure about it. I have gone through its documentation but still not able to run it properly. I request you to please help me in this. Regards, Gunjan On Wed, Nov 19, 2014 at 6:50 PM, Bruce Fischl <fischl@nmr.mgh.harvard.edu> wrote: Hi Gunjan you need to give us more details. What are you trying to do? Why not just run recon-all -all? That will be much easier than trying to ru individual steps chers Bruce On Wed, 19 Nov 2014, Gunjan Gautam wrote: Hi, I am facing few issues. I will be thankful if you can help me out. Query 1) I used "mri_convert a.nii a.resliced.nii -rl b.nii" in order to generate slices so that comparison may take place properly. The command ran properly but generating cropped slices. I am attaching the input volume, reference volume and the cropped slice screenshot. [icon_10_generic_list.png] a.nii [icon_10_generic_list.png] b.nii Query 2) For skull stripping I ran below command bibl04@BIBL04:~/Desktop/freesurfer/bin$ recon-all -i a.nii -autorecon1 and its showing error for subject id. I am really not aware, what to write. ERROR: must specify a subject id Query 3) I did not find that which option from (1 to 31 ) should generate the "binary brain mask". Thanks, Gunjan On Wed, Nov 19, 2014 at 12:41 AM, Gunjan Gautam <gunjan.gemini29@gmail.com> wrote: Ok, I go through that. Thanks, Gunjan On Nov 19, 2014 12:37 AM, "Bruce Fischl" <fischl@nmr.mgh.harvard.edu> wrote: Hi Gunjan, it's all on the wiki. You should go through our tutorials cheers Bruce On Wed, 19 Nov 2014, Gunjan Gautam wrote: Thank you I will be thankful if you could let me know about the link related to recon-all, in order to know about the parameters(input volume info etc) associated to it and other usage. On Nov 19, 2014 12:18 AM, "Bruce Fischl" <fischl@nmr.mgh.harvard.edu> wrote: Hi Gunjan just run recon-all and it will do everything that we know how to do (including those steps) cheers Bruce On Wed, 19 Nov 2014, Gunjan Gautam wrote: Ok, I try this tomorrow morning. I will be thankful if you could tell me about getting brain mask and performing skull stripping otherwise I'll have to wait for 12 hours to get your replies as our working hours doesn't match. On Nov 18, 2014 8:45 PM, "Z K" <zkaufman@nmr.mgh.harvard.edu> wrote: Gunjan, this may be related to an Ubuntu specific issue we address on the Freesurfer download page: https://surfer.nmr.mgh.harvard.edu/fswiki/Download Specifically the part thats states the following: *** On Ubuntu, you might have to do this if qdec or freeview complain about libjpeg or libtiff: cd /usr/lib (/usr/lib/i386 or /usr/lib/x86_64 on some Ubuntu distros) sudo ln -s libjpeg.so.8 libjpeg.so.62 sudo ln -s libtiff.so.4 libtiff.so.3 *** -Zeke On 11/18/2014 03:55 AM, Gunjan Gautam wrote: > Hello Sir, > > I have place the "license.txt" in the same directory that contains > build-stamp.txt as there were no directory with name "FREESURFER_HOME". > freeview command is not working and showing below message > "freeview.bin: error while loading shared libraries: libjpeg.so.62: > cannot open shared object file: No such file or directory". > What next I need to do in order to generate a output brain mask. > > Best Regards, > Gunjan > > > > On Thu, Nov 13, 2014 at 12:24 AM, Z K <zkaufman@nmr.mgh.harvard.edu > <mailto:zkaufman@nmr.mgh.harvard.edu>> wrote: > > Im having a little difficulty following all the avenues of your attempts > to get this working, but the bottom line is that the error message below > is complaining that you do not have a valid license file. > > You need to go to this page and register and get the valid license file: > > https://surfer.nmr.mgh.harvard.edu/registration.html > > Once you get a valid license file you will need to place it in your > $FREESURFER_HOME directory (the same directory that contains > build-stamp.txt) and give it the name .license or license.txt. Try both. > > -Zeke > > On 11/12/2014 01:29 PM, Gunjan Gautam wrote: > > Sir earlier I was trying to use FreeSurfer using Virtual Box in my > > Windows machine. There the mri_convert command did not work so I > > switched to a Linux machine. Now, I'm using centos6_x86_64 version as my > > machine has the Ubuntu 12.04 platform and here also facing the same > > problem. > > > > I want to share one thing that in my institute I tried hard to download > > FreeSurfer but every time it resulted into a time-out situation so I had > > to ask someone to get it downloaded and send me through Google drive. > > This was how I installed it in my system. > > > > Best regards, > > Gunjan > > > > On Nov 12, 2014 9:15 PM, "Z K" <zkaufman@nmr.mgh.harvard.edu <mailto:zkaufman@nmr.mgh.harvard.edu> > > <mailto:zkaufman@nmr.mgh.harvard.edu <mailto:zkaufman@nmr.mgh.harvard.edu>>> > wrote: > > > > Gunjan, > > > > Are you using your own version of VirtualBox, with freesurfer installed > > on it by you? Or the VirtualBox version of freesurfer made by us and > > downloaded from our website? I am confused because in your email below, > > freesurfer appears to be installed on the Desktop whereas the > > prepackaged version made by us is
_______________________________________________ Freesurfer mailing list Freesurfer@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.
...