Hi Leonor,
I find that using the combination of SNR/CNR to generate a summary QC metric and identify potentially problematic subjects works best most of the times.
These measures quantify the quality of the results obtained after the pre-processing stage. E.g., if a subject moved a lot and has been flagged as an outlier in most of the motion-related estimated parameters but EDDY has successfully corrected the data, the SNR and CNR will pass the threshold (see Discussion in the paper).
However, if you have a subject which is ‘borderline’ ok and the other metrics point to potential issues, it is definitely worth looking at the summary maps stored in the single subject report generated by eddy_quad and/or at the whole corrected dataset itself.
Hope this helps, cheers,
Matteo
> On 22 Mar 2019, at 08:15, Leonor Agan <[log in to unmask]> wrote:
>
> Hi Matteo,
>
> What about those subjects who 'passed' the SNR/CNR metric but failed (i.e., outliers) in all the other metrics? Do you exclude those?
>
> Thanks!
> Leonor
>
> ########################################################################
>
> To unsubscribe from the FSL list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|