unprocessed diffusion data

69 views
Skip to first unread message

batel yifrah

unread,
Oct 22, 2023, 4:21:42 PM10/22/23
to HCP-Users
Hello experts,

I'm looking to preprocess the diffusion data on my own, but I've encountered some confusion when dealing with the unprocessed data of the subjects. Within this data, I've come across numerous NIfTI files named 'dir95,' 'dir96,' 'dir97,' and 'SBRef.' I'm uncertain about why I need images for directions 95, 96, and 97, rather than just one of them. Additionally, I'm puzzled about the purpose of the 'SBRef' image.

Furthermore, I initially believed that running a top-up required a complete set of phase encoding images, and only one image necessary for the phase encoding reversed direction. So, why do I have, for instance, 96 images for RL and 96 for LR, instead of 96 RL images and just 1 LR image?

Any assistance or clarification you can provide would be greatly appreciated.

Thank you,
Batel

Tim Coalson

unread,
Oct 24, 2023, 12:09:32 AM10/24/23
to hcp-...@humanconnectome.org
The diffusion directions were acquired in separate sequences, I believe, and we choose different lengths for each run to reduce the chance of pairing a diffusion file with the wrong directions file.

For diffusion, we use eddy for preprocessing, because the way I understand it, the diffusion gradients cause additional distortion through induced eddy currents.  So, I believe we used the same diffusion gradient direction+polarity with opposite phase encoding to make it easier to estimate these extra distortions.

Tim


--
You received this message because you are subscribed to the Google Groups "HCP-Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hcp-users+...@humanconnectome.org.
To view this discussion on the web visit https://groups.google.com/a/humanconnectome.org/d/msgid/hcp-users/30a4dbb1-b8cb-4e62-be91-f0445877532an%40humanconnectome.org.

Tim Coalson

unread,
Oct 24, 2023, 12:14:36 AM10/24/23
to hcp-...@humanconnectome.org
Oh, and the diffusion gradient direction causes localized changes to the contrast (that is basically the point), so having the same diffusion direction with the opposite phase is a more direct image matching problem than going to SBRef or some other difusion direction.

Tim

batel yifrah

unread,
Oct 24, 2023, 3:29:03 AM10/24/23
to HCP-Users, HCP-Users
Thank you very much! So, should I run eddy separately for each run? As for the XTRACT(by FSL), how should I go about combining the results?

thanks,
Batel

Harms, Michael

unread,
Oct 24, 2023, 11:18:56 AM10/24/23
to hcp-...@humanconnectome.org, tim.c...@gmail.com

 

No, for maximum quality you should run ‘eddy’ on all 6 runs of dMRI simultaneously.  This is all handled seamlessly if you process with the HCPpipelines.

 

Stepping back, if you are using the HCP-YA data, why do you need to preprocess the dMRI data “on your own” in the first place?

 

Cheers,

-MH

 

-- 

Michael Harms, Ph.D.

-----------------------------------------------------------

Professor of Psychiatry

Washington University School of Medicine

Department of Psychiatry, Box 8134

660 South Euclid Ave.                        Tel: 314-747-6173

St. Louis, MO  63110                          Email: mha...@wustl.edu

 

From: batel yifrah <batel....@gmail.com>
Reply-To: "hcp-...@humanconnectome.org" <hcp-...@humanconnectome.org>
Date: Tuesday, October 24, 2023 at 2:29 AM
To: HCP-Users <hcp-...@humanconnectome.org>
Cc: "tim.c...@gmail.com" <tim.c...@gmail.com>
Subject: Re: [hcp-users] unprocessed diffusion data

 

* External Email - Caution *

 


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.

Harms, Michael

unread,
Oct 24, 2023, 11:23:09 AM10/24/23
to hcp-...@humanconnectome.org, tim.c...@gmail.com

 

Also, please see the 2013 paper describing the choices behind the HCP-YA dMRI protocol:

https://www.ncbi.nlm.nih.gov/pubmed/23702418

 

 

 

-- 

Michael Harms, Ph.D.

-----------------------------------------------------------

Professor of Psychiatry

Washington University School of Medicine

Department of Psychiatry, Box 8134

660 South Euclid Ave.                        Tel: 314-747-6173

St. Louis, MO  63110                          Email: mha...@wustl.edu

 

Glasser, Matt

unread,
Nov 1, 2023, 4:57:11 AM11/1/23
to hcp-...@humanconnectome.org, tim.c...@gmail.com

I would ignore anyone one who say, “you have to run denoising on raw data.”  These folks don’t understand that modern accelerated MR reconstructions have already done far more to the noise profiles than does distortion and motion correction to the same image grid.

 

Matt.

Reply all
Reply to author
Forward
0 new messages