External Email - Use Caution
Regarding alternatives: you could run something like mriqc (https://pypi.org/project/mriqc/) on the source nifti/dicom files. It includes SNR measures if I remember correctly.
Tim
On August 7, 2020 at 4:22 PM "Douglas N. Greve" dgreve@mgh.harvard.edu wrote:
It should be related to the SNR, but including it as a noise measure regressor in a GLM sounds dicey to me.
On 8/6/2020 5:02 AM, Scheffler, F, Mev [fredas@sun.ac.za] wrote:
External Email - Use Caution
Dear FreeSurfer Devs
I got a comment about including a SNR variable in my main stats analysis and so came across a Rosen et al 2018 paper which refers to the Euler number. I then checked the archives and came across a thread https://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg58748.html where you say that in version 6.0 the numbers are included in the aseg.stats file, specifically as a global measure called SurfaceHoles.
With regards to our QC procedures: We visually inspected all of our scans and then followed ENIGMA’s quality assessment protocols. We excluded scans with motion artefacts or poor-quality images and only included the data from the scans with acceptable or good quality segmentation. Outlier detection and the distribution of our data was also checked prior to analysis of the data.
So, my question is, can I include the SurfaceHoles variable in my model as a proxy for SNR? Would that be sensible, or do you have an alternative suggestion?
Thanks very much!
Kind regards,
Freda Scheffler
PhD Candidate
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://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
-- Dr. Tim Schäfer Postdoc Computational Neuroimaging Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy University Hospital Frankfurt, Goethe University Frankfurt am Main, Germany