Hi
There appears to be an issue with writing the output from the GMM fit to the list of mixtur modl parameters. This should only rarely happen as a switch to Gaussian mixtu modeling typically indicates a model order which is too high (too many components) as an intermediate fix I suggest to switch off the mixture modeling ( using the --no_mm option ) and/or limiting the model order.
Also, please feel free to upload the melodic_ic file and melodic_mix so we can track down where mixture modeling fails.
Hth
Christian
Sent from my iPad
On 31.10.2012, at 19:57, David Drake <[log in to unmask]> wrote:
> Dear Experts,
>
> I have been experiencing similar problems with melodic as previously posted. Our lab does most of their pre-processing in AFNI, and we are new to ICA in general. I've been trying to conduct resting state analysis on already processed resting data that was registered to standard space (Talairach). After running my melodic command line script, I was getting that same error:
>
> An exception has been thrown
> Logic error:- detected by Newmat: index error: requested index = 0
>
> MatrixType = Rect # Rows = 1; # Cols = 3
>
> I originally was just going to exclude the subjects that didn't work, but I tried rearranging my subjects on the subject list which proved to work some of the time. I would like to include all of my subjects.
>
> Any help with this problem would be greatly appreciated!
>
> I uploaded a melodic_mix and melodic_IC files for this run, and the reference number is 682548.
>
> Here is the script i was using:
> ***************
> home=/data/drakedm/stor0/drake/FSL_ICA
> subject_list2=/data/drakedm/stor0/drake/subject_list_david2.txt
>
> cd ${home}
>
> melodic --approach=concat \
> --in=${subject_list2} \
> --outdir=${home}/ICA_singlecat \
> --report \
> --tr=1.9 \
> --Oall \
> --verbose
>
> *******************
> the subject_list_david2.txt is a list of the pre-processed datasets:
> PDC003-1_resting_preICA.nii.gz
> PDC004-1_resting_preICA.nii.gz
> PDC006-1_resting_preICA.nii.gz
> PDC011-1_resting_preICA.nii.gz
> PDC013-1_resting_preICA.nii.gz
> PDC014-1_resting_preICA.nii.gz
> PDP003-1_resting_preICA.nii.gz
> PDP006-2_resting_preICA.nii.gz
> PDP010-1_resting_preICA.nii.gz
> PDP013-1_resting_preICA.nii.gz
> PDP018-1_resting_preICA.nii.gz
>
> part of the log file before script terminated:
> ********************
> IC map 26 ...
> calculating mixture-model fit
> try Gaussian Mixture Model
> saving probability map: /data/drakedm/stor0/drake/FSL_ICA/ICA_singlecat/stats/probmap_26
> re-scaling spatial maps ...
> thresholding ...
> alternative hypothesis test at p > 0.5
> saving thresholded Z-stats image: /data/drakedm/stor0/drake/FSL_ICA/ICA_singlecat/stats/thresh_zstat26
> creating report page ... done
> IC map 27 ...
> calculating mixture-model fit
> try Gaussian Mixture Model
>
> Thank you and Cheers,
>
> David
|