Hi Stephen,
After using data without motion correction and remove slice mean, now
I can use Melodic Multi-session temporal concatenation with 6 runs
from the same subject.
However, the result is weird. Below are the links of the output of 2 components
and registration report of one of the run. I use default setting from
the Melodic gui for Data, Pre-stats, Registration and Post-stat
http://www.flickr.com/photos/40507517@N06/4907933869/
http://www.flickr.com/photos/40507517@N06/4907933747/
http://www.flickr.com/photos/40507517@N06/4907933795/
I also have 2 questions,
1: in the registration output, there are red lines align outside of
the brain. Is this cause the ICA output do not align with the
background image? What are those red lines stand for?
2: after running Melodic Multi-session temporal concatenation, the
result melodic_IC.nii.gz is in subject space or reference space?
Thank you very much for your help!
Sincerely,
Isabella
On Wed, Aug 18, 2010 at 4:42 AM, Stephen Smith <[log in to unmask]> wrote:
> Hi - this still sounds really odd - like there's a problem with running out
> of RAM/swap, or something weird in your data.
> What resolution is the concatenated data?
> Try converting the data to floats, also look carefully at the concatenated
> timeseries in FSLView. The line
>
> /usr/local/fsl/bin/fslstats prefiltered_func_data -k mask -p 50
> 0.000000
>
> suggests that something is problematic in the data you're feeding in…..?
> Cheers.
>
> On 17 Aug 2010, at 15:41, Isabella Murphy wrote:
>
> Hi Stephen,
>
> I have prefiltered_func_data_smooth in
> ~/fsl-4.1.6-feeds/data/fmri.feat/prefiltered_func_data_smooth.nii.gz
>
> I can view it with FSLView
>
> when I run fslstats prefiltered_func_data_smooth -r -R, I got 0.000000
> 8756.932617 0.000000 14402.849609
>
> If I do the same 2 runs as single-session ICA, no problem.
>
> Also, if I concatenate the same 2 runs outside of FSL using other
> software (Bioimagesuite) then run in Melodic using single-session ICA,
> turn off all the pre-stats options,uncheck everything in Registration
> tab, uncheck Threshold IC maps in Post-Stats tab, it will not work
> neither:
>
> Progress Report / Log
> Started at Tue Aug 17 10:38:57 EDT 2010
>
> Main script
>
> /bin/cp /tmp/feat_2dgDV8.fsf design.fsf
>
> mkdir .files;cp /usr/local/fsl/doc/fsl.css .files;ln -s
> /usr/local/fsl/doc/images .files/images
>
> /usr/local/fsl/bin/fsl_sub -T 45 -l logs -N feat2_pre
> /usr/local/fsl/bin/feat
> /Users/administrator/Desktop/concatenated2runs.ica/design.fsf -D
> /Users/administrator/Desktop/concatenated2runs.ica -I 1 -prestats
> 6056
>
> Prestats
>
> /usr/local/fsl/bin/fslmaths
> /Users/administrator/testdata/fastICA/tr6381_bis_intrinput
> prefiltered_func_data -odt float
> Total original volumes = 458
>
> /usr/local/fsl/bin/fslroi prefiltered_func_data example_func 229 1
>
> /usr/local/fsl/bin/fslstats prefiltered_func_data -p 2 -p 98
> -1.247978 1.251370
>
> /usr/local/fsl/bin/fslmaths prefiltered_func_data -thr -0.9980432
> -Tmin -bin mask -odt char
>
> /usr/local/fsl/bin/fslstats prefiltered_func_data -k mask -p 50
> 0.000000
>
> /usr/local/fsl/bin/fslmaths mask -dilF mask
>
> /usr/local/fsl/bin/fslmaths prefiltered_func_data -mas mask
> prefiltered_func_data_thresh
> divide by zero
> while executing
> "expr $normmean / $median_intensity "
> (procedure "feat5:proc_prestats" line 406)
> invoked from within
> "feat5:proc_prestats $session "
> ("-prestats" arm line 2)
> invoked from within
> "switch -- [ lindex $argv $argindex ] {
>
> -I {
> incr argindex 1
> set session [ lindex $argv $argindex ]
> }
>
> -D {
> incr argindex 1
> set..."
> ("for" body line 2)
> invoked from within
> "for { set argindex 1 } { $argindex < $argc } { incr argindex 1 } {
> switch -- [ lindex $argv $argindex ] {
>
> -I {
> incr argindex 1
> set ses..."
> (file "/usr/local/fsl/bin/feat" line 137)
>
>
>
> I have OS X 10.5.8, FSL build 416
> Thank you very much!
> Isabella
>
> On Tue, Aug 17, 2010 at 3:01 AM, Stephen Smith <[log in to unmask]> wrote:
>
> Hi - something is going wrong in the middle of the preprocessing, which is
>
> very strange. Can you find the image file:
>
> prefiltered_func_data_smooth
>
> and view it is FSLView and also run
>
> fslstats prefiltered_func_data_smooth -r -R
>
> This would suggest to me that maybe you ran out of disk space or quota
>
> during the analysis?
>
> Cheers.
>
> On 16 Aug 2010, at 16:54, Isabella Murphy wrote:
>
> Hi Stephen and Harry,
>
> I have difficulty to use MELODIC with Multi-session temporal concatenation
>
> as well. I tried to give 2 runs of the same subject, I also tried to give 1
>
> run of 2 different subjects. Both seems to have some problem at registration
>
> stage. I tried to turn off all the pre-processing options and do not provide
>
> any image for registration, the result still the same. It seems something
>
> wrong at the registration stage (the registration results missing 1 image)
>
> and there is no ICA component calculated.
>
> Below is a log after I do 2 runs of 1 subject using Multi-session temporal
>
> concatenation. I am running MELODIC 3.10 on mac 10.5. Thank you for your
>
> help.
>
> Sincerely,
>
> Isabella Murphy
>
> Programming Assistant II
>
> Department of Diagnostic Radiology
>
> Yale University
>
>
> Progress Report / Log
>
> Started at Thu Aug 12 17:00:36 EDT 2010
>
> Main script
>
> /bin/cp /tmp/feat_0vh6E9.fsf design.fsf
>
> mkdir .files;cp /usr/local/fsl/doc/fsl.css .files;ln -s
>
> /usr/local/fsl/doc/images .files/images
>
> /usr/local/fsl/bin/fsl_sub -T 45 -l logs -N feat2_pre
>
> /usr/local/fsl/bin/feat /Users/administrator/Desktop/2runs.ica/design.fsf
>
> -D /Users/administrator/Desktop/2runs.ica -I 1 -prestats
>
> 14541
>
>
> Prestats
>
> /usr/local/fsl/bin/fslmaths
>
> /Users/administrator/testdata/fastICA/tr6381_bis_intrinput
>
> prefiltered_func_data -odt float
>
> Total original volumes = 458
>
> /usr/local/fsl/bin/fslroi prefiltered_func_data example_func 229 1
>
> /usr/local/fsl/bin/fslmaths prefiltered_func_data -Tmean mean_func
>
> /usr/local/fsl/bin/bet2 mean_func mask -f 0.3 -n -m; /usr/local/fsl/bin/immv
>
> mask_mask mask
>
> /usr/local/fsl/bin/fslmaths prefiltered_func_data -mas mask
>
> prefiltered_func_data_bet
>
> /usr/local/fsl/bin/fslstats prefiltered_func_data_bet -p 2 -p 98
>
> -1.247978 1.251370
>
> /usr/local/fsl/bin/fslmaths prefiltered_func_data_bet -thr -0.9980432 -Tmin
>
> -bin mask -odt char
>
> /usr/local/fsl/bin/fslstats prefiltered_func_data -k mask -p 50
>
> 0.000000
>
> /usr/local/fsl/bin/fslmaths mask -dilF mask
>
> /usr/local/fsl/bin/fslmaths prefiltered_func_data -mas mask
>
> prefiltered_func_data_thresh
>
> /usr/local/fsl/bin/fslmaths prefiltered_func_data_thresh -Tmean mean_func
>
> /usr/local/fsl/bin/susan prefiltered_func_data_thresh 0.9359835
>
> 2.12314225053 3 1 1 mean_func 0.9359835 prefiltered_func_data_smooth
>
> /usr/local/fsl/bin/fslmaths prefiltered_func_data_smooth -mas mask
>
> prefiltered_func_data_smooth
>
> ** ERROR (nifti_image_read): failed to find header file for
>
> 'prefiltered_func_data_smooth'
>
> ** ERROR: nifti_image_open(prefiltered_func_data_smooth): bad header info
>
> Error: failed to open file prefiltered_func_data_smooth
>
> Cannot open volume prefiltered_func_data_smooth for reading!
>
> divide by zero
>
> while executing
>
> "expr $normmean / $median_intensity "
>
> (procedure "feat5:proc_prestats" line 406)
>
> invoked from within
>
> "feat5:proc_prestats $session "
>
> ("-prestats" arm line 2)
>
> invoked from within
>
> "switch -- [ lindex $argv $argindex ] {
>
> -I {
>
> incr argindex 1
>
> set session [ lindex $argv $argindex ]
>
> }
>
> -D {
>
> incr argindex 1
>
> set..."
>
> ("for" body line 2)
>
> invoked from within
>
> "for { set argindex 1 } { $argindex < $argc } { incr argindex 1 } {
>
> switch -- [ lindex $argv $argindex ] {
>
> -I {
>
> incr argindex 1
>
> set ses..."
>
> (file "/usr/local/fsl/bin/feat" line 137)
>
>
>
> ---------------------------------------------------------------------------
>
> Stephen M. Smith, Professor of Biomedical Engineering
>
> Associate Director, Oxford University FMRIB Centre
>
> FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
>
> +44 (0) 1865 222726 (fax 222717)
>
> [log in to unmask] http://www.fmrib.ox.ac.uk/~steve
>
> ---------------------------------------------------------------------------
>
>
>
>
>
>
>
> ---------------------------------------------------------------------------
> Stephen M. Smith, Professor of Biomedical Engineering
> Associate Director, Oxford University FMRIB Centre
>
> FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
> +44 (0) 1865 222726 (fax 222717)
> [log in to unmask] http://www.fmrib.ox.ac.uk/~steve
> ---------------------------------------------------------------------------
>
>
>
>
|
