help
488 Subscribers 
help > Loading conn files in SPM
Showing 26-30 of 30 posts
Nov 9, 2017 03:11 PM | Ines Del Cerro - IDIBELL - Bellvitge Biomedical Research Institute
RE: Loading conn files in SPM
Hi all,
This may be useful: https://www.jiscmail.ac.uk/cgi-bin/webad...
Just replace the script on SPM folder.
Best,
Ines
This may be useful: https://www.jiscmail.ac.uk/cgi-bin/webad...
Just replace the script on SPM folder.
Best,
Ines
Sep 14, 2018 09:09 PM | laurentcohen - ICM
RE: Loading conn files in SPM
Hello
I have the same problem exactly.
I exported the filtered data from CONN and entered them in a SPM model. During estimation I received the message:
Failed 'Model estimation'
Error using spm_est_non_sphericity (line 208)
Please check your data: There are no significant voxels.
In file "/home/imen/github/spm12/spm_est_non_sphericity.m" (v6913), function "spm_est_non_sphericity" at line 208.
In file "/home/imen/github/spm12/spm_spm.m" (v7120), function "spm_spm" at line 433.
In file "/home/imen/github/spm12/config/spm_run_fmri_est.m" (v7057), function "spm_run_fmri_est" at line 34.
The 2015 thread has been interrupted before a solution was found... Do you have any suggestion?
Thank you very much in advance
Laurent
Originally posted by Alfonso Nieto-Castanon:
I have the same problem exactly.
I exported the filtered data from CONN and entered them in a SPM model. During estimation I received the message:
Failed 'Model estimation'
Error using spm_est_non_sphericity (line 208)
Please check your data: There are no significant voxels.
In file "/home/imen/github/spm12/spm_est_non_sphericity.m" (v6913), function "spm_est_non_sphericity" at line 208.
In file "/home/imen/github/spm12/spm_spm.m" (v7120), function "spm_spm" at line 433.
In file "/home/imen/github/spm12/config/spm_run_fmri_est.m" (v7057), function "spm_run_fmri_est" at line 34.
The 2015 thread has been interrupted before a solution was found... Do you have any suggestion?
Thank you very much in advance
Laurent
Originally posted by Alfonso Nieto-Castanon:
Hi
Chaleece,
This indicates that SPM finds no significant "any condition/task effects" in your data. There are several potential reasons, my best guess is that this is related to the explicit masking of the analyses combined with the band-pass filtering of your original data (by default SPM will look for "in-brain" voxels by comparing their average BOLD signal to the average global BOLD signal; since your data is band-pass filtered the average BOLD signal is always 0). I would suggest to change in your script the line "spm.stats.fmri_spec.mthresh=.8" to ""spm.stats.fmri_spec.mthresh=-inf" in order to skip this implicit masking altogether. Please let me know if that does not seem to be the issue here.
Best
Alfonso
Originally posted by Chaleece Sandberg:
This indicates that SPM finds no significant "any condition/task effects" in your data. There are several potential reasons, my best guess is that this is related to the explicit masking of the analyses combined with the band-pass filtering of your original data (by default SPM will look for "in-brain" voxels by comparing their average BOLD signal to the average global BOLD signal; since your data is band-pass filtered the average BOLD signal is always 0). I would suggest to change in your script the line "spm.stats.fmri_spec.mthresh=.8" to ""spm.stats.fmri_spec.mthresh=-inf" in order to skip this implicit masking altogether. Please let me know if that does not seem to be the issue here.
Best
Alfonso
Originally posted by Chaleece Sandberg:
Hello again,
I hate to be a bother, but I have now run into another problem when I try to actually run the GLM in SPM:
Running 'Model estimation'
SPM12: spm_spm (v6015) 11:58:27 - 06/08/2015
========================================================================
SPM12: spm_est_non_sphericity (v6015) 11:58:30 - 06/08/2015
========================================================================
Chunk 97/97 : ...processing
Failed 'Model estimation'
Error using spm_est_non_sphericity (line 196)
Please check your data: There are no significant voxels.
In file "C:\Users\Public\Documents\spm12\spm_est_non_sphericity.m" (v6015), function "spm_est_non_sphericity" at line 196.
In file "C:\Users\Public\Documents\spm12\spm_spm.m" (v6015), function "spm_spm" at line 418.
In file "C:\Users\Public\Documents\spm12\config\spm_run_fmri_est.m" (v5809), function "spm_run_fmri_est" at line 33.
The following modules did not run:
Failed: Model estimation
Does this mean that the Denoising step somehow reduced any significance? I didn't include the conditions as confounds for denoising. I am attaching my my batch for spm, if that helps. Would you like the conn.mat file?
I hate to be a bother, but I have now run into another problem when I try to actually run the GLM in SPM:
Running 'Model estimation'
SPM12: spm_spm (v6015) 11:58:27 - 06/08/2015
========================================================================
SPM12: spm_est_non_sphericity (v6015) 11:58:30 - 06/08/2015
========================================================================
Chunk 97/97 : ...processing
Failed 'Model estimation'
Error using spm_est_non_sphericity (line 196)
Please check your data: There are no significant voxels.
In file "C:\Users\Public\Documents\spm12\spm_est_non_sphericity.m" (v6015), function "spm_est_non_sphericity" at line 196.
In file "C:\Users\Public\Documents\spm12\spm_spm.m" (v6015), function "spm_spm" at line 418.
In file "C:\Users\Public\Documents\spm12\config\spm_run_fmri_est.m" (v5809), function "spm_run_fmri_est" at line 33.
The following modules did not run:
Failed: Model estimation
Does this mean that the Denoising step somehow reduced any significance? I didn't include the conditions as confounds for denoising. I am attaching my my batch for spm, if that helps. Would you like the conn.mat file?
Dec 19, 2018 02:12 PM | Laila Franke
RE: Loading conn files in SPM
Hi both,
Were you able to find an answer to your question?
Thanks in advance!
Best, laila
Were you able to find an answer to your question?
Thanks in advance!
Best, laila
Dec 19, 2018 02:12 PM | Pravesh Parekh - National Institute of Mental Health and Neurosciences
RE: Loading conn files in SPM
Hello,
You can use the Import utility in the GUI to achieve what you want. Go to "Project" and click on Import. Then you can select a SPM.mat file. Conn will automatically import the functional images and experimental design from the SPM.mat file along with the realignment (motion) parameters as a first level covariate. You might have to specify the structural files manually (but if you use the "Find" button, it is very efficient and requires minimal effort, as long as you correctly specify the filters).
As an alternate, you could specify the smoothed functional files as primary data set and the normalized functional files as the secondary data set. Normalized skull stripped images are usually specified in the structural files tab but it does not matter too much (structural files are only used for display purposes during denoising and first level; if they are normalized, you should be fine). You would have to specify the rp_*.txt file for each subject as a first level covariate.
I would suggest going about the first method as it i much more convenient. Do check your overall experiment design and other parameters before proceeding further (for example when importing SPM.mat file into a new project, you might have to delete the "rest" condition).
Finally, you might want to run the outlier detection from preprocessing tab before proceeding to denoising.
Hope this helps
Best
Pravesh
You can use the Import utility in the GUI to achieve what you want. Go to "Project" and click on Import. Then you can select a SPM.mat file. Conn will automatically import the functional images and experimental design from the SPM.mat file along with the realignment (motion) parameters as a first level covariate. You might have to specify the structural files manually (but if you use the "Find" button, it is very efficient and requires minimal effort, as long as you correctly specify the filters).
As an alternate, you could specify the smoothed functional files as primary data set and the normalized functional files as the secondary data set. Normalized skull stripped images are usually specified in the structural files tab but it does not matter too much (structural files are only used for display purposes during denoising and first level; if they are normalized, you should be fine). You would have to specify the rp_*.txt file for each subject as a first level covariate.
I would suggest going about the first method as it i much more convenient. Do check your overall experiment design and other parameters before proceeding further (for example when importing SPM.mat file into a new project, you might have to delete the "rest" condition).
Finally, you might want to run the outlier detection from preprocessing tab before proceeding to denoising.
Hope this helps
Best
Pravesh