Print

Print
Thanks, Steve and Taylor. The example I sent was with the default settings in FEAT, so it was BBR for example_func2highres. I tested that again with DOF set to 7 and it does look better but still a bit distorted (https://www.dropbox.com/s/bw8ekd07wseh5eb/ABCD-data_BadRegistration_7DOF.png?dl=0). I also took a closer look at the input data, and it seems distorted/compressed there, so it must be an issue with the image (e.g., wrong subject) or perhaps their distortion correction caused a problem. I'm trying to track down the raw data for this scan to see if something went awry with their processing pipeline...

Their pipeline is a little opaque, but maybe they will eventually release data that meets the BIDS standard. We've had some success doing the conversion to BIDS with some of the raw data using heudiconv; but last time I checked, it was only working with the Siemens data so we switched back to their minimally processed data.

Thanks!
David


On Wed, Oct 30, 2019 at 1:58 PM Taylor Hanayik <[log in to unmask]> wrote:
Hi David, 

Another thing to try is to use the single band reference image for the func to highres registration. It’ll have much better SNR than the multi band ABCD EPI “example_func”. Not sure if this is done or not in their pipeline. There’s no mention of it in https://www.biorxiv.org/content/biorxiv/early/2018/11/04/457739.full.pdf




Cheers,

Taylor Hanayik


Taylor Hanayik PhD
Analysis Research Software Engineer
FMRIB, John Radcliffe Hospital
University of Oxford
[log in to unmask]





On 30 Oct 2019, at 17:29, David V. Smith <[log in to unmask]> wrote:

Hello,

I'm working on some analyses with the ABCD data, but I'm running into a registration issue with some of the data. We're using their minimally preprocessed data, which I believe includes B0 distortion correction and motion correction.


In the example above (which is just using FLIRT), I'm assuming the mismatch between the example_func and standard is due to the mismatch between the example_func and the highres. But, it's not clear to me whether/how the mismatch between example_func and highres can be corrected. It almost looks as if the fieldmap was applied incorrectly. Any thoughts/advice?

If they can't be corrected, should we exclude these types of subjects from our analyses? 

Thanks!
David

--
David V. Smith, Ph.D.
Assistant Professor of Psychology
Temple University
Weiss Hall, Room 825
1701 North 13th Street
Philadelphia, PA 19122
Office Phone: 215-204-1552

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



--
David V. Smith, Ph.D.
Assistant Professor of Psychology
Temple University
Weiss Hall, Room 825
1701 North 13th Street
Philadelphia, PA 19122
Office Phone: 215-204-1552

To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1