JISCMail - FSL Archives

On Fri, Apr 25, 2014 at 4:17 PM, Mark Jenkinson <[log in to unmask]> wrote:

Hi,

The problem is right at the beginning of your analysis.

Your registrations have failed.

It is really important to go through the process of checking (i.e. looking at) your outputs, as the wiki page and the FSL Course practical both explain, before proceeding to the later steps.

I suspect that your registration is having problems because of the large amount of neck and initial orientation of your images. This can probably be solved by running fslreorient2std and robustfov on your images before doing any of the steps for FIRST. So try that and hopefully it will fix your problems, but remember to always look at your data and the outputs from each analysis step.

All the best,

Mark

On 25 Apr 2014, at 11:41, R B <[log in to unmask]>

wrote:

Dear Mark,
I have attached all the files as suggested, do i have to attach the .bvars &.vtk files of other subcortical structures too. kindly let me know about it.

i did check the specified folder i found these .bmv files to be precise, i have BrStem_bin, L_Accu_bin, L_Amyg_bin, L_caud_bin, L_Hipp_bin, L_Late_bin, R_Accu_bin, R_Amyg_bin, R_Caud_bin, R_Hip_bin, R_Late_bin

thanks in advance for your valuable support

Regards

Hsekar

On Fri, Apr 25, 2014 at 4:01 PM, Mark Jenkinson <[log in to unmask]> wrote:

Hi,

I need to see all the output of run_first_all from one subject, not just the bvars file.

So please send me the vtk and the nifti (nii.gz) files as well.

Also, since I now realise that you are running on Ubuntu, can you check that the directory $FSLDIR/data/first/models_336_bin/ contains many .bmv files and three directories. I'm asking about this because sometimes on Ubuntu systems people can fail to install all the necessary files necessary for FIRST to run.

All the best,

Mark

On 25 Apr 2014, at 11:18, R B <[log in to unmask]> wrote:

Dear Mark,
i have uploaded the files at the specified link,

output of the run_first_all : FIRST_C_1-L_Hipp_first.bvars

output after concatenating the bvar files is : L_Hipp_all.bvars

design matrix: design.mat

kindly provide ur guidance in proceeding.

on trying "--useReconMNI" i get the error message

read model

done reading model

setting up shape/appearance model

The shape has 732 vertices.

336 modes of variation are retained.

The model was constructed from 336 training subjects.

model loaded

terminate called after throwing an instance of 'NEWMAT::ConvergenceException'

but the fsl closes unexpectedly, with the error message, i have attached the photoshoot of error message.

thank you

On Fri, Apr 25, 2014 at 3:10 PM, Mark Jenkinson <[log in to unmask]> wrote:

Hi,

This sounds fine, although something is clearly going wrong given the huge values in your output (e.g. 2.0072e+17 = 2.0072*10^17). I assume that you have checked that the outputs from run_first_all look OK, and that you have created the shape_analysis directory (for your outputs). If so, then I think we need to see some files to understand what is going wrong. Send us the following (single subject output from run_first_all, the output of cancat_bvars and the design.mat). You can send the files to the following site:

https://oxfile.ox.ac.uk/oxfile/work/extBox?id=68312615463381F4C

Also, it might be worth trying an analysis with --useReconMNI rather than --useReconNative and --useRigidAlign, just in case there is a problem with the content or location of the individual registration matrices.

All the best,

Mark

On 25 Apr 2014, at 10:31, R B <[log in to unmask]>

wrote:

Thank U Mark for replying
here goes the commands that i have used

run_first_all -i Con_01 -o First_Con_01, similarly i did for Con_02, Con_03, Pat_01, Pat_02, Pat_03,

so i had files First_Con_01_*.bvars files

since i was interested in shape analysis of hippocampus

i used concat_bvars FIRST_L_Hippo_all First_Con_01_L_Hipp_first.bvars .........Con_02.........

so i had the concatenated file FIRST_L_Hippo_all.bvars

i created a design.mat fle, it was just a 2 group design

and then i did the command first_utils

first_utils --vertexAnalysis --usebvars -i FIRST_L_Hippo_all.bvars -d design.mat -o shape_analysis/con1_dis2_L_Hipp --useReconNative --useRigidAlign -v >& shape_analysis/C_P_L_Hipp_output.txt

all these analysis were analysed in the same computer, in the last 3-4 days. and i get the error message.

as described earlier

thank u in advance for your valuable support

regards

Hsekar

On Fri, Apr 25, 2014 at 2:06 PM, Mark Jenkinson <[log in to unmask]> wrote:

Hi,

What is the exact command that you are running?

And have you run all the preceding steps yourself recently?

All the best,

Mark

On 24 Apr 2014, at 11:55, R B <[log in to unmask]> wrote:

Dear Bergsland,
thanks a ton,

i had stuck at this place for weeks,

but in the next step of first_utils

i was shown this error message

read model

done reading model

setting up shape/appearance model

The shape has 732 vertices.

336 modes of variation are retained.

The model was constructed from 336 training subjects.

model loaded

read model

done reading model

setting up shape/appearance model

The shape has 732 vertices.

336 modes of variation are retained.

The model was constructed from 336 training subjects.

2.0072e+17 881.051 5.25073e-05 0.000188217

48.5903 6.41596e-10 2.23193e+17 1.556e+17

2.34483e+17 14093.8 2.32187e+17 0.0125242

2.0072e+17 14092.6 38.0579 48.5899

14096.8 5.25072e-05 13.7711 48.59

55.0677 48.59 0.0124105 789.439

48.5899 2.00702e+17 2.24266e+17 3.67338e-29

194.36 2.44923e+38 55.0588 3.82395e+17

12.1476 1.99576e+17 0.00340571 4.8715e+13

2.22094e+17 1.9495e+26 2.20981e+17 13.7662

1.84873e+17 2.31096e+17 220.208 48.59

5.46824e+16 0.00074484 13.7687 2.64842e+17

3173.77 1.52252e+17 55.0559 48.5901

1.82875e-07 55.058 5.25668e-05 9.38417e+18

2.00702e+17 1.68046e+17 14096.8 7.31122e+11

4.72724e-05 0.0108808 2.22055e+17 2.51367e+17

2.19842e+17 0.0124627 2.11688e+17 220.259

12.3975 194.361 -2.18716e+17 3523.67

2.81376e-09 -7.65347e+25 0.000225766 55.0587

-0.00599172 4.19081e-17 3.47812e-07 48.59

13.764 5.25678e-05 48.59 2.07462e+17

5.2498e-05 2.83983e+17 2.09722e+17 2.08583e+17

2.17511e+17 5.43975e+16 880.925 1.54597e+17

2.82892e+17 5.2497e-05 5.25064e-05 781.441

0.0210916 2.4462e-15 220.235 3526.14

48.5901 0.023384 4.52531e-08 2.67103e+17

881.02 0.000210025 1.51091e+17 1.14264e+10

1.99572e+17 55.0657 2.9558e-06 5.25075e-05

46136.4 2.08544e+17 220.259 13.774

-0.00156579 55.0581 0.00017582 12.1475

0.000209969 12.1475 2.33655e+26 12.1475

2.28876e+17 5.87024e-19 48.5903 12.1476

done recon

terminate called after throwing an instance of 'NEWMAT::ConvergenceException'

can you pl. guide me what could have possibly gone wrong this error messae

On Thu, Apr 24, 2014 at 1:16 PM, Niels Bergsland <[log in to unmask]> wrote:

What output do you get when you run head -n3 on the file generated by concat_bvars ?

Also, are you really seeing "usr/share/fsl/5.0" rather than "/usr/share/fsl/5.0" ?

On Thu, Apr 24, 2014 at 9:28 AM, R B <[log in to unmask]> wrote:

Dear Bergsland
Thanks for your reply,

so if my understand is correct, after running the command run_first_all on individual subjects i get the .bvars file for each of the subcortical structures.

now as i am interested in the shape analysis of left hippocampus, do i have to edit the file " subj01_L_hippo_first.bvars "

in that case i the path mentioned in this file is usr/share/fsl/5.0.

or did you mention the bvars file generated after the command concat_bvars

in this case there is no mention about the destination "usr/share/fsl/5.0"

can u kindly please tell me whether i have mis-understood at any step

thanks in advance

regards

hsekar

On Wed, Apr 23, 2014 at 7:30 PM, Niels Bergsland <[log in to unmask]> wrote:

The bvars is created when you run FIRST. So you'll find them in the same directory as your other outputs.

On Wed, Apr 23, 2014 at 3:30 PM, Hsekar B <[log in to unmask]> wrote:

dear fsl experts
while performing FSL- FIRST analysis i encountered this problem
after completing the first_run_all and concat_bvars commands
i was able to get through these steps without any trouble and obtained the output as described in the manual.
however on initiating the command first_utilis i get the following error message.

Warning: does not exist. Attempting to switch to: /usr/share/fsl/5.0
read model
Not a vtk file (error in line 1).
Not a vtk file (error in line 1).

in a link (below) it has been discussed to
Load the bvar file in a text editor and change anything that looks like /usr/local/fsl with the path for your own installation of FSL (which might not be /usr/share/fsl/5.0).

https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind1308&L=fsl&D=0&P=381683

can anyone kindly let me know where can i get this bvar file,

thanks in advance

regards

RB

--
Niels Bergsland
Integration Director
Buffalo Neuroimaging Analysis Center
100 High St. Buffalo NY 14203
[log in to unmask]

--
Niels Bergsland
Integration Director
Buffalo Neuroimaging Analysis Center
100 High St. Buffalo NY 14203
[log in to unmask]

<C_P_L_Hipp_output.txt>