Hi Matt
I uploaded the scripts that I use to this folder
<https://caltech.box.com/s/u6bqsq6itv99k5pbf1n7lifmtz4eqjih> [
https://caltech.box.com/s/u6bqsq6itv99k5pbf1n7lifmtz4eqjih] (I call
VolumeProcessingPipelineBatch.sh, which calls
GenericfMRIVolumeProcessingPipeline_mcflirt.sh, which calls among others
MotionCorrection_MCFLIRTbased.sh -- recall that I made small alterations to
the preprocessing pipeline to do MC with MCFLIRT).
I also uploaded some data for one problematic subject: the unprocessed
volumes (subfolder "unprocessed"), and the processed folder for one
resting-state run ("rsBOLDMB1"). The EPI data is 2.5mm isotropic, 1s TR,
multiband acceleration factor 4 -- the protocol is also uploaded
("rsBOLD_MB.pdf").
Thanks for any insights,
- Julien
PS1: You'll notice that the T2 for that subject is very low res (we were
not yet acquiring a high res T2) - I imagine this is suboptimal for the
pipeline, but that is what I have to work with (hopefully it does not
completely throw things off)
PS2: If you have other comments on the outputs of the preprocessing, other
than the AP flip which is obviously problematic, let me know. I personally
find that the MNI warped EPIs look quite noisy... and also there are weird
bumps in the FNIRT MNI registration of the structural scans. I could of
course wait for the HCP course to discuss all this in person, but since I
have your attention now...
Julien Dubois
Postdoctoral Scholar
California Institute of Technology, Pasadena, CA
http://www.emotion.caltech.edu/~jdubois
On Fri, Apr 17, 2015 at 4:19 PM, Glasser, Matthew <[email protected]>
wrote:
> Can you give us more details on the data you are feeding into the
> pipelines and how you called fMRIVolume? We have not experienced this
> issue with HCP data or the non-HCP data we have tried the pipelines with.
>
> Peace,
>
> Matt.
>
> From: Julien Dubois <[email protected]>
> Date: Friday, April 17, 2015 at 4:13 PM
> To: "[email protected]" <[email protected]>
> Subject: [HCP-Users] Minimal Preprocessing Pipeline (on my own data):
> Flipped EPI (AP)
>
> Dear developers
>
> I ran the minimal preprocessing pipeline on a number of my own subjects,
> and recently noticed that there was a rather unfortunate AP flip of the EPI
> during registration for ~30% of the subjects (explaining some rather odd
> results that I have been trying to track down for a couple of days).
>
> The first volume in which I can notice this AP flip is the
> "Scout_gdc_undistorted_fieldmap2str.nii.gz" in
> "${SUBJECT_DIR}/${RUN}/DistortionCorrectionAndEPIToT1wReg_FLIRTBBRAndFreeSurferBBRbased/"
> (that the first first showing the incorrect AP orientation, according to
> `ls -lat`) . I'm having a hard time tracking down in the code
> (DistortionCorrectionAndEPIToT1wReg_FLIRTBBRAndFreeSurferBBRbased.sh) where
> this volume is created/used, and hence how to remedy the bug.
>
> Let me know if you can help. Thanks!
> - Julien
>
> _______________________________________________
> HCP-Users mailing list
> [email protected]
> http://lists.humanconnectome.org/mailman/listinfo/hcp-users
>
>
> ------------------------------
>
> The materials in this message are private and may contain Protected
> Healthcare Information or other information of a sensitive nature. If you
> are not the intended recipient, be advised that any unauthorized use,
> disclosure, copying or the taking of any action in reliance on the contents
> of this information is strictly prohibited. If you have received this email
> in error, please immediately notify the sender via telephone or return mail.
>
_______________________________________________
HCP-Users mailing list
[email protected]
http://lists.humanconnectome.org/mailman/listinfo/hcp-users
