Dear Alfonso, dear all,

is there an easy way to compute FD Jenkinson using the first level realignment covariate for use as a 2nd level covariate in CONN?

Thank you!
Best wishes

Julian

Dear Julian,

Please try the attached patch using the syntax:

   conn_convertl12l1covariate FD_jenkinson

That will create in your CONN project a new 'QC_FDjenkinson' covariate with the Framewise Displacement timeseries (using Jenkinson definition and FSL's defaults, instead of CONN's). See "help conn_convertl12l1covariate" for additional options (the script allows you to use different Sphere radius or center definitions in Jenkinson's FD formulation, use CONN's or Power's formulation instead, etc.) This script is still under development so please let me know if you run into any issues

Best
Alfonso
Originally posted by Julian Dronse:

Dear Alfonso, dear all,

I came across this thread and ran the patch, which worked great - I got the FD Power values I was looking for. Thanks so much for this.

However, after running the patch, only the FD variable is listed in the 1st level Covariates; all other covariates are gone (e.g. realignment, scrubbing, QC timeseries). Has anyone noticed this before?

Is there a way to import the missing variables back into the 1st level Covariates, to avoid having to rerun the preprocessing steps? Which are the files that would need to be imported?

Thanks so much for any help you can provide!

Best wishes,

Madeleine


Originally posted by Alfonso Nieto-Castanon:

Dear Madeleine,

Sorry about that, my impression is that the behavior you describe is due to having only uncompressed one of the two files in the patch. To fix this you may simply re-load your CONN project (since any changes performed by this script to your project will not have not been saved yet), then make sure to unzip and copy both files (named conn_convertl12l2covariates.m and conn_module.m) to your conn distribution folder, and then try again. 

Having said that, this patch was really for CONN's release 18a, and in the latest release (18b) this functionality is already part of the main distribution. If you update to 18b, then from the main CONN gui, in the Setup.Covariates.FirstLevel tab, go to the 'covariate tools' menu and select there the option named 'compute new/derived first-level covariates' to find the option that creates FD_jenkinson measures directly without the need for any patch. 

Hope this helps
Alfonso
Originally posted by madeleine: