External Email - Use Caution
Hello, that's a follow up regarding my issue. Do you have any suggestions? Please find enclosed the X.mat file from selxavg3 error (ill conditioned design)
I will test using different parameters. For both expanding ring and rotating wedge 1cycle will last about 50 seconds and each stimulus will have 6 cycles (now I had 1 cycle lasting 200 sec). I guess the error is with timing; I tried to change the duration of 1cycle while doing some test analysis and the commands were performed without errors, so I suppose using the new settings the computation will work.
Thank you for your insight. Marco
On Tue, 21 Jan 2020 at 16:58, Marco Ninghetto m.ninghetto@nencki.edu.pl wrote:
Thank you for your reply. I think I've better explained the situation in a further email, which I paste in here.
I used expanding ring and rotating wedge. For my pilot, unfortunately, we used only 2 cycles for each stimulus and 1 cycle lasted 200 seconds (for sure I will change it to have more cycles, lasting less). TR = 2.5.
Retinotopy paradigm files are as follow: -for expanding ring ("pos" direction since it expands): stimtype eccen direction pos -for rotating wedge ("neg" direction since is counter-clockwise): stimtype polar direction neg
To compute analyses, I ran the following pipeline:
- recon-all
- preproc-sess
- mkanalysis-sess
- selxavg3-sess
Having only two cycles for each stimulus, in mkanalysis-sess, I used the option -hpf and lowered the filter to 0.5.
In the last level (selxavg3-sess), I get the error. Basically this step runs up to when MATLAB starts to calculate. Here I get the following error:
MATLAB is selecting SOFTWARE OPENGL rendering. < M A T L A B (R) > Copyright 1984-2019 The MathWorks, Inc. R2019b Update 3 (9.7.0.1261785) 64-bit (glnxa64) November 27, 2019
To get started, type doc. For product information, visit www.mathworks.com.
> >> /usr/local/freesurfer/fsfast/toolbox/fast_selxavg3.m
/usr/local/freesurfer/fsfast/toolbox/fast_ldanaflac.m /usr/local/freesurfer/matlab/MRIread.m
> starting fast_selxavg3b
#@# SUBJ02 ############################### /media/sf_w_Desktop/w_pRF/retinotopy/SUBJ02
$Id: fast_selxavg3b.m,v 1.4 2016/05/04 22:16:47 greve Exp $
outtop = /media/sf_w_Desktop/w_pRF/retinotopy Extension format = nii.gz INFO: key nSliceGroups unrecognized, line 11, skipping nruns = 2 autostimdur =
outanadir = /media/sf_w_Desktop/w_pRF/retinotopy/SUBJ02/bold/retinotopy.lh Found 118266/125623 (94.1) voxels in mask 1 Creating Design Matrix ... creation time = 0.008 sec DoMCFit = 1 ntptot = 260, nX = 158, DOF = 102 Saving X matrix to /media/sf_w_Desktop/w_pRF/retinotopy/SUBJ02/bold/retinotopy.lh/Xtmp.mat XCond = 6.85663e+16 (normalized) ERROR: design is ill-conditioned ans = 1
ERROR: fast_selxavg3() failed\n
I tried to modify, out of curiosity, the duration of 1 cycle lower than 100 seconds in "mkanalysis-sess". It made the job. So I assume that the problem is with that parameter. For further subjects I'll use different timings. But is there still a way to compute calculations on these stimuli? Moreover, now I acquire only one functional img for each stimulus. Do you recommend me to do more runs? (maybe having only 1 run causes the issue).
Thank you for your help! Marco
On Tue, 21 Jan 2020, 16:16 Greve, Douglas N.,Ph.D., < DGREVE@mgh.harvard.edu> wrote:
Actually, I was able to get those files from your previous emails
On 1/15/2020 11:24 AM, Marco Ninghetto wrote:
External Email - Use CautionSorry for the late reply, the message was non stop bounced.
So, here You will find:
- terminal output for `selxavg3-sess` (filename: terminal_output.txt). I
copied only the last part when freesurfer accesses to MATLAB and where the error occurs;
- the mat file from expanding ring and for rotating wedge (filename:
expring_SUBJ02_20191122.mat and rotwedge_SUBJ02_20191122.mat). https://drive.google.com/open?id=1vkSRHeTPEn-UE4iighsStYbNkrzn03PG
- paradigm files for expanding ring and for rotating wedge
Thank you for your precious help, Marco
On Fri, 3 Jan 2020 at 17:51, Greve, Douglas N.,Ph.D. < DGREVE@mgh.harvard.edu> wrote:
Can you send your paradigm files? Can you also send the terminal output? One more thing, can you send the X.mat file from the output folder?
On 1/2/2020 3:19 AM, Marco Ninghetto wrote:
External Email - Use CautionHello experts, I'm performing retinotopy analysis but when running the command
selxavg3-sessIt starts, but after some lines, I get the following error:
ERROR: design is ill-conditionedI don't know how to proceed from here to solve the error. My retinotopy stimuli make 2 cycles (e.g.: rotating wedge makes 2 full cycles before quit) and each cycle takes 200 seconds to complete. TR is set to 2.5 seconds.
Please, does anyone have any advice? Thanks in advance, Marco
--
Marco Ninghetto, PhD candidate Laboratory of Neuroplasticity Nencki Institute of Experimental Biology Polish Academy of Sciences 3 Pasteur Street, 02-093 Warsaw, Poland
Freesurfer mailing listFreesurfer@nmr.mgh.harvard.eduhttps://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
--
Marco Ninghetto, PhD candidate Laboratory of Neuroplasticity Nencki Institute of Experimental Biology Polish Academy of Sciences 3 Pasteur Street, 02-093 Warsaw, Poland
Freesurfer mailing listFreesurfer@nmr.mgh.harvard.eduhttps://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer