Dear all:
I also observed similar situations for the data sets being
realigned, normalized and smoothed from SPM.
But after I redo the analysis with original data set (
no pre-processing). The results came out. I guess that
the pre-processing was "destroying" the characteristics
if the signals. Or at least, smoothing out the signals.
Another question: where can I find the standard method to select
the resulted IC maps from Melodic. Based on different papers
of De Luca (2006), or Greicius (2004), some kind of spatial cross
correlation or best fit was used. How to do these test?
Can we just use visual inspection?
Any comments is appreciated.
Wen
On Oct 19, 2007, at 3:04 AM, Christian F. Beckmann wrote:
> Hi,
>
> did you maybe run out of disk space? The estimation completes and
> it does enter the code segment which writes results to disk - I
> can't think of anything at that point which could go wrong in our i/
> o code. so my guess would be that the system write call did not
> complete. I suspect there is no 'melodic finished' message in the
> log.txt? If not, does the binary continue to be run (what happens
> if you check with 'top'?)
> hth
> Christian
>
> On 19 Oct 2007, at 05:42, Zhen Zonglei wrote:
>
>> Hi, all:
>> When I used MELODIC ICA to do multi-sessions resting state fMRI
>> analysis (three session's temporal concatenation,200 time points
>> every session and delete first 3 volume) with FSL4.0.1, the
>> program seemed run completely. But, no ICs were shown in 'high
>> level MELODIC report' and only 'Input' and 'registration summary'
>> were displayed. In addition, I can find 'melodic_IC.nii.gz' and
>> other file in groupmelodic.ica directory.
>> Can someone tell me what the problem is?
>>
>> Please help me.
>> Zonglei Zhen
>>
>> The following is MELODIC logs:
>> Melodic Version 3.05
>> /nfs/h1/r1/jliu/new_soft_test/freesurfer/fsl/bin/melodic -
>> i .filelist -o groupmelodic.ica -v --nobet --bgthreshold=3 --
>> tr=3.0 --report --guireport=../../report.html --bgimage=bg_image --
>> mmthresh= 0.5 -a concat
>> ---------------------------------------------
>> Melodic results will be in groupmelodic.ica
>> Create mask ... done
>> Reading data file rest/
>> gefunc1281420200RS2007080801LJHuangYis005a001_results.ica/
>> reg_standard/filtered_func_data ... done
>> Estimating data smoothness ... done
>> Removing mean image ... done
>> Reading data file rest/
>> gefunc1281420200RS2007081601LJHuangYis005a001_results.ica/
>> reg_standard/filtered_func_data ... done
>> Removing mean image ... done
>> Reading data file rest/
>> gefunc1281420200RS2007091401LJhuangyis005a001_results.ica/
>> reg_standard/filtered_func_data ... done
>> Removing mean image ... done
>> Excluding voxels with constant value ... done
>> Normalising by voxel-wise variance ... done
>>
>> Initial data size : 197 x 23485
>>
>> Reading data file rest/
>> gefunc1281420200RS2007080801LJHuangYis005a001_results.ica/
>> reg_standard/filtered_func_data ... done
>> Removing mean image ... done
>> Reading data file rest/
>> gefunc1281420200RS2007081601LJHuangYis005a001_results.ica/
>> reg_standard/filtered_func_data ... done
>> Removing mean image ... done
>> Reading data file rest/
>> gefunc1281420200RS2007091401LJhuangyis005a001_results.ica/
>> reg_standard/filtered_func_data ... done
>> Removing mean image ... done
>>
>> Data size : 219 x 23485
>>
>> Starting PCA ... done
>> Start whitening using 73 dimensions ...
>> retaining 83.8176 percent of the variability
>> ... done
>>
>> Starting ICA estimation using concat
>>
>> Step no. 1 change : 0.993451
>> Step no. 2 change : 0.377832
>> ………………………………………….
>> Step no. 45 change : 0.000605192
>> Step no. 46 change : 0.000475788
>> Convergence after 46 steps
>>
>> Sorting IC maps
>> Calculating T- and S-modes
>>
>> Writing results to :
>> groupmelodic.ica/melodic_IC
>> groupmelodic.ica/melodic_Tmodes
>> groupmelodic.ica/melodic_Smodes
>> groupmelodic.ica/melodic_mix
>> groupmelodic.ica/melodic_FTmix
>> groupmelodic.ica /melodic_PPCA
>> groupmelodic.ica/melodic_ICstats
>> groupmelodic.ica/mask
Wen-Ching Liu, Ph.D.
Assistant Professor
Functional Imaging Laboratory
Department of Radiology
University of Medicine and Dentistry
of New Jersey-New Jersey Medical School
973-972-2281/fax: 973-972-0836
http://njms.umdnj.edu/departments/radiology/fmri/index.cfm
|