Re: [HCP-Users] Clarification in change in fMRI reconstruction algorithm

[Harms, Michael]

Hi Ben,
The following replies to your questions are from Steen Moeller (cc’ed) at CMRR.

1/ the reconstruction for all HCP data is performed using the slice-grappa algorithm. There was a change in Q3 that improved on the implementation of it relative to some low-level sequence detail concerning
 the use of the blipping scheme in blipped-CAIPI reference data. The choosen algorithm however has stayed the same. The SpSg algorithm has not been used, but is available in new versions of the sequence.
2/ the individual coil data are not available. If this is only needed for a single session, that can then be made available for a representative dataset. If so then email me directly to make this available.
cheers,



-- 
Michael Harms, Ph.D.

---
Conte Center for the Neuroscience of Mental Disorders
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:  on behalf of Benjamin Risk 
Date: Monday, May 9, 2016 at 10:41 PM
To: "Glasser, Matthew" 
Cc: "hcp-users@humanconnectome.org" 
Subject: Re: [HCP-Users] Clarification in change in fMRI reconstruction algorithm





They didn't use in-plane GRAPPA but they used a multi-band variant on GRAPPA. Setsompop 2012 is called Slice-GRAPPA. I am asking about the details of the multiband reconstruction.


Thanks,
Ben


On Mon, May 9, 2016 at 10:51 PM, Glasser, Matthew 
 wrote:


fMRI didn’t use GRAPPA, rather it was an update to the multi-band reconstruction.


Peace,


Matt.




From:  on behalf of Benjamin Risk 
Date: Monday, May 9, 2016 at 12:55 PM
To: "hcp-users@humanconnectome.org" 
Subject: [HCP-Users] Clarification in change in fMRI reconstruction algorithm







Dear HCP experts,


I have two questions:


1) I understand that the reconstruction algorithm for fMRI changed with the Q3 data release. I believe there are three main GRAPPA reconstruction approaches for multiband imaging: SENSE/GRAPPA from Blaimer et al (2006); Slice-GRAPPA from Setsompop et al
 (2012); and Split-Slice-GRAPPA from Cauley et al (2014). Do the HCP reconstruction algorithms correspond to these methods? From my reading of Ugurbil et al 2013, p. 89, I thought the change was from SENSE/GRAPPA to Slice-GRAPPA, but I now see that the overview
 document http://www.cmrr.umn.edu/multiband/Multi-Band_C2P_Instructions_R013a.pdf includes a citation to Cauley.


2) I ultimately want to simulate data generated by coil sensitivities similar to the HCP scanners and head coils. Are single-band calibration data for each coil available? 




Thank you,
Ben Risk

Postdoc at SAMSI / Department of Biostatistics UNC Chapel Hill



___
HCP-Users mailing list
HCP-Users@humanconnectome.org
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
HCP-Users@humanconnectome.org
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 listHCP-Users@humanconnectome.orghttp://lists.humanconnectome.org/mailman/listinfo/hcp-users

Re: [HCP-Users] Question about myelin maps in HCP S900 data

[Glasser, Matthew]

Always use MSMAll data for group analysis.  I recommend BC for neuroanatomy, 
and without BC for quantitative analysis.  You can always smooth a myelin map, 
but I recommend no smoothing.

Peace,

Matt.

From: 
>
 on behalf of Aaron C >
Date: Wednesday, May 11, 2016 at 1:35 AM
To: "hcp-users@humanconnectome.org" 
>
Subject: [HCP-Users] Question about myelin maps in HCP S900 data

Dear HCP experts,

I have some questions about the myelin maps in HCP S900 distribution. In the 
folder 100307\MNINonLinear\fsaverage_LR32k, I see there are seven *.dscalar.nii 
files of myelin maps:

100307.MyelinMap.32k_fs_LR.dscalar.nii
100307.MyelinMap_BC.32k_fs_LR.dscalar.nii
100307.MyelinMap_BC_MSMAll.32k_fs_LR.dscalar.nii
100307.MyelinMap_MSMAll.32k_fs_LR.dscalar.nii
100307.SmoothedMyelinMap.32k_fs_LR.dscalar.nii
100307.SmoothedMyelinMap_BC.32k_fs_LR.dscalar.nii
100307.SmoothedMyelinMap_BC_MSMAll.32k_fs_LR.dscalar.nii

>From these names, I guess the files with "BC" are residual bias field 
>corrected myelin maps, and the files with "MSMAll" are MSMAll-registered 
>myelin maps. Am I correct?

For any group analysis, should I use MSMAll-registered myelin map from each 
individual for better registration between subjects?

Thank you.

___
HCP-Users mailing list
HCP-Users@humanconnectome.org
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
HCP-Users@humanconnectome.org
http://lists.humanconnectome.org/mailman/listinfo/hcp-users

[HCP-Users] Question about myelin maps in HCP S900 data

[Aaron C]

Dear HCP experts,

I have some questions about the myelin maps in HCP S900 distribution. In the 
folder 100307\MNINonLinear\fsaverage_LR32k, I see there are seven *.dscalar.nii 
files of myelin maps:

100307.MyelinMap.32k_fs_LR.dscalar.nii
100307.MyelinMap_BC.32k_fs_LR.dscalar.nii
100307.MyelinMap_BC_MSMAll.32k_fs_LR.dscalar.nii
100307.MyelinMap_MSMAll.32k_fs_LR.dscalar.nii
100307.SmoothedMyelinMap.32k_fs_LR.dscalar.nii
100307.SmoothedMyelinMap_BC.32k_fs_LR.dscalar.nii
100307.SmoothedMyelinMap_BC_MSMAll.32k_fs_LR.dscalar.nii

>From these names, I guess the files with “BC” are residual bias field 
>corrected myelin maps, and the files with “MSMAll” are MSMAll-registered 
>myelin maps. Am I correct?
 
For any group analysis, should I use MSMAll-registered myelin map from each 
individual for better registration between subjects?

Thank you.
  
___
HCP-Users mailing list
HCP-Users@humanconnectome.org
http://lists.humanconnectome.org/mailman/listinfo/hcp-users

Re: [HCP-Users] A question about saving custom parcellation label map in a CIFTI *.dlabel.nii file

[Aaron C]

This works great! Thank you!
From: glass...@wustl.edu
To: aaroncr...@outlook.com; hcp-users@humanconnectome.org
Subject: Re: [HCP-Users] A question about saving custom parcellation label map 
in a CIFTI *.dlabel.nii file
Date: Sun, 8 May 2016 19:13:07 +






Save the matrix data as dscalar and then make the label table as an 
appropriately formatted text file and combine them with wb_command 
-cifti-label-import.



Peace,



Matt.





From:  on behalf of Aaron C 


Date: Sunday, May 8, 2016 at 2:07 PM

To: "hcp-users@humanconnectome.org" 

Subject: [HCP-Users] A question about saving custom parcellation label map in a 
CIFTI *.dlabel.nii file








Dear HCP experts,



I have a question about saving my custom parcellation label map in a CIFTI 
*.dlabel.nii file. I can now save the labels in a *.dscalar.nii file by using 
the CIFTI header of an existing *.dscalar.nii file, but I cannot do the same 
for saving my labels in *.dlabel.nii
 file directly by using the CIFTI header of an existing *.dlabel.nii file,  
because I need to use my own label table for defining label color scheme 
different from the one used in the CIFTI header of the existing *.dlabel.nii 
file.



Would you please advise how can I create a custom *.dlabel.nii file with my own 
ROI names, key values, and corresponding RGB color values from a vector of 
labels for each vertex (32k mesh)?



Many thanks!



___

HCP-Users mailing list

HCP-Users@humanconnectome.org

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
HCP-Users@humanconnectome.org
http://lists.humanconnectome.org/mailman/listinfo/hcp-users