Registration to MNI space

[hsp...@g.rit.edu]

Hello Frank,

I have QA maps obtained using GQI reconstruction. I want to map these QA maps to the MNI space such that the resulting QA maps in MNI space are equivalent to QA maps obtained using QSDR reconstruction. How can I do this using DSI-Studio? 

Note that I cannot directly run QSDR since I do not have access to the DWI files.  

Thank you,

Harsh.

[Frank Yeh]

GUI:
[Tools][O7:Nonlinear Registration Toolbox]->select
\template\HCP1065.QA.nii.gz as the template image, and subject QA map
as the subject image

Command line: http://dsi-studio.labsolver.org/Manual/command-line-for-dsi-studio#TOC-Rigid-body-rotation-and-spatial-normalization
--action=reg

If any of these does not work, please let me know.
Frank

> --
> You received this message because you are subscribed to the Google Groups "DSI Studio" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to dsi-studio+...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/19ced1cc-954b-4214-b5f0-85cda446b822n%40googlegroups.com.

[Harsh Prajapati]

Hello Frank,

I used the command line option and it works but the R2 value I get after registration is around 0.51 even after tuning the parameters. I tried this on a test DWI file I have. Running QSDR on this DWI file results in a R2 value of around 0.64, but if I run GQI first on the same DWI file and then register the QA map obtained to the HCP1065 QA template using the instructions you provided in the previous email, I get an R2 values of around 0.51. Also, the range of QA values after registration is [-0.4346,14.]

> To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/CAG_Qrt%2BkuXHwwXy5APF_d2tDtW459H_QyrSKPPhOfu8LOqYFbA%40mail.gmail.com.

[Harsh Prajapati]

Continuing the previous email.

Also, the range of QA values after registration is [-0.4346,14.0466]. How do you normalize these values between [0,1] specifically how do you handle the negative values?

Thank You,
Harsh.

[Frank Yeh]

QSDR uses not only QA map but also ISO map for registration, so the
accuracy is much better.
If you have an ISO map, then you may add it (the documentation to see
if it improves the R2 value.

The QA ranges seem to be wrong. I can help check your data to
troubleshoot the issue (if you can upload the images and the command
you used using the dropbox upload link).

Best regards,
Frank

> To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/641B499C-1541-46AB-8417-429CB74EBBF6%40g.rit.edu.

[Harsh Prajapati]

Hello Frank,

Sorry I forgot to mention, the R2 value of 0.51 is after using ISO map for registration. I have uploaded the DWI file and the corresponding bval and bevc files on the dropbox.

The command I used for registration is here below:

 ./dsi_studio.app/Contents/MacOS/dsi_studio --action=reg --from=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz.nqa0.nii.gz --from2=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz.iso.nii.gz --to=/Users/harshprajapati/Desktop/dsi_studio.app/Contents/MacOS/template/HCP1065.QA.nii.gz --to2=/Users/harshprajapati/Desktop/dsi_studio.app/Contents/MacOS/template/HCP1065.ISO.nii.gz --constraint=0.5 --smoothness=0.1 --iteration=64

Thank You,

Harsh.

To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/CAG_Qrt%2BKsPxDOUi%3DP8ybpLaTVNDTY5sQ-BGdw8JrPDPEpqMhWQ%40mail.gmail.com.

[Frank Yeh]

I will check and get back to you.
Best regards,
Frank

> To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/4C5B7E8C-A859-4BF5-B7F2-A74217C06E07%40g.rit.edu.

[Harsh Prajapati]

Ok, Thanks a lot.

> You received this message because you are subscribed to a topic in the Google Groups "DSI Studio" group.
> To unsubscribe from this topic, visit https://groups.google.com/d/topic/dsi-studio/kytdlVQgmY8/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to dsi-studio+...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/CAG_QrtL8b%3DStdc_tCaY_SB66tkr90k3Z7WwPegBzp%2By2O23Nsg%40mail.gmail.com.

[Frank Yeh]

Here's my test results:

command:

dsi_studio --action=reg --from=qa.nii.gz --from2=iso.nii.gz
--to=HCP1065.QA.nii.gz --to2=HCP1065.ISO.nii.gz > log.txt

output:
DSI Studio Jun 9 2021, Fang-Cheng Yeh
action=reg
from=qa.nii.gz
to=HCP1065.QA.nii.gz
from2=iso.nii.gz
to2=HCP1065.ISO.nii.gz
output=qa.nii.gz.wp.nii.gz
running linear registration.
reg_type=1
0.496 0.0188925 -0.00433917 24.0056
-0.00979805 0.499157 0.0228945 26.6759
0.00191708 -0.0226582 0.451521 10.7415
correlation cofficient: 0.715306
reg_type=1
normalize_signal=1
normalizing signals
nonlinear registration using dual image modalities
resolution=2
smoothness=0.3
constraint=0.5
iteration=60
min_dimension=32
R2: 0.658109
output warpped image: qa.nii.gz.wp.nii.gz

> To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/4C5B7E8C-A859-4BF5-B7F2-A74217C06E07%40g.rit.edu.

[Harsh Prajapati]

Hello Frank,

These results look good. I wonder why am I getting lower R2 values? Can you tell me the command you used to perform GQI reconstruction and also the version of DSI-studio used?

Thank You,
Harsh.

> To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/CAG_QrtL-%3DpR00Amgy6epQqJ7sGpHeisdodCLos4tcW212MqHLQ%40mail.gmail.com.

[Frank Yeh]

I use all default settings (GQI with 1.25 sampling length ratio).
The version is June 9 2021, but I think the difference is the
parameters you specified (e.g. smoothing)
Frank

> To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/17DFE556-E230-4161-84B4-187D7DEF9566%40g.rit.edu.

[Harsh Prajapati]

Hello Frank,

I had tried with the same parameters values you used and got even lower R2 value of 0.47. I tried downloading the latest version of DSI-studio using the One Drive link for Mac OS but it still downloads the Jun 5 build and the Google Drive link is not working it throws a 404 error.

Thank You,
Harsh.

> To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/CAG_QrtJWZcRcEC%2B%2BLTmpXDrkmbz9df8yA5VsmTvXv75Pg0SFdg%40mail.gmail.com.

[Frank Yeh]

I am sorry about the Mac download issue (fixing it), but the Jun 5
should give the same result.

Could you show me the exact command you used and the log output?

I will test on my side, and sorry about the issue.

Best,
Frank

> To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/A8B2B5AD-9108-46FF-9E2F-58AE5BFC4B9C%40g.rit.edu.

[Harsh Prajapati]

Hello Frank,

Here are the list of commands and their output log :

1. Create src file

./dsi_studio.app/Contents/MacOS/dsi_studio --action=src --source=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz --bval=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/series_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6.bval --bvec=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/series_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6.bvec > log_src.txt

output log:

DSI Studio Jun  5 2021, Fang-Cheng Yeh

source=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz

action=src

loading nifti

bval=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/series_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6.bval

bvec=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/series_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6.bvec

flip_by=1

flip_by=1

output src to /Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz

up_sampling=0

sort_b_table=0

Save Files

2. GQI reconstruction:

./dsi_studio.app/Contents/MacOS/dsi_studio --action=rec --source=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz --method=4 --param0=1.25 > log_rec.txt

Output log:

DSI Studio Jun  5 2021, Fang-Cheng Yeh

source=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz

action=rec

loading source...

src loaded

method=4

param0=1.25

odf_order=8

odf_resolving=0

record_odf=0

dti_no_high_b=1

check_btable=1

other_output=fa,nqa,rdi,nrdi

num_fiber=5

r2_weighted=0

thread_count=4

half_sphere=0

scheme_balance=1

motion_correction=0

start reconstruction...

checking b-table

012=BEST,012fx=-8%,012fy=-5%,012fz=-6%,021=-8%,021fx=-8%,021fy=-9%,021fz=-11%

102=-7%,102fx=-9%,102fy=-10%,102fz=-6%,120=-12%,120fx=-13%,120fy=-12%,120fz=-13%

210=-5%,210fx=-10%,210fy=-5%,210fz=-10%,201=-8%,201fx=-7%,201fy=-8%,201fz=-7%

GQI reconstruction

saving pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz

reconstruction finished.

3. Export nqa0 and ISO :

./dsi_studio.app/Contents/MacOS/dsi_studio --action=exp --source=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz --export=nqa0,iso > log_exp.txt

Output log:

DSI Studio Jun  5 2021, Fang-Cheng Yeh

source=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz

action=exp

export=nqa0,iso

loading /Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz...

retriving matrix nqa0

write to file /Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz.nqa0.nii.gz

retriving matrix iso

write to file /Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz.iso.nii.gz

4. Dual mode registration to MNI space:

./dsi_studio.app/Contents/MacOS/dsi_studio --action=reg --from=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz.nqa0.nii.gz --from2=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz.iso.nii.gz --to=/Users/harshprajapati/Desktop/dsi_studio.app/Contents/MacOS/template/HCP1065.QA.nii.gz --to2=/Users/harshprajapati/Desktop/dsi_studio.app/Contents/MacOS/template/HCP1065.ISO.nii.gz > log_reg.txt

Output log:

DSI Studio Jun  5 2021, Fang-Cheng Yeh

action=reg

from=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz.nqa0.nii.gz

to=/Users/harshprajapati/Desktop/dsi_studio.app/Contents/MacOS/template/HCP1065.QA.nii.gz

from2=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz.iso.nii.gz

to2=/Users/harshprajapati/Desktop/dsi_studio.app/Contents/MacOS/template/HCP1065.ISO.nii.gz

output=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz.nqa0.nii.gz.wp.nii.gz

running linear registration.

reg_type=1

0.510241 0.0325879 0.00670809 18.7643

-0.0171856 0.362992 -0.37997 40.9939

-0.0176401 0.271581 0.355306 -10.1915

correlation cofficient: 0.547662

reg_type=1

normalize_signal=1

normalizing signals

nonlinear registration using dual image modalities

resolution=2

smoothness=0.3

constraint=0.5

iteration=60

min_dimension=32

R2: 0.491514

output warpped image: /Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz.nqa0.nii.gz.wp.nii.gz

Thank You,

Harsh.

To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/CAG_QrtLQ7XtcrsBjWaLdZGnBs63h-s4bPuNVa%2BLWwRLT9%2BxMiw%40mail.gmail.com.

[Frank Yeh]

Hi Harsh,

Thank you for your help. I have identified the problem/bug in
--action=exp where DSI Studio failed to output the correct srow matrix
to the nifti file.

I have updated DSI Studio (today's version) to fix the problem.

Please update DSI Studio and see if it works for you.

Best regards,
Frank

> To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/407DD341-CE95-46A5-A29B-46D0DE8E5B23%40g.rit.edu.

[Harsh Prajapati]

Hello Frank,

I downloaded today’s version from the box link for MacOS and ran the export command but it throws me the following error:

ERROR: cannot find nqaO in /Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_02/EMBC21_base_Time_01/pair_5.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20111115155840_5_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz

Thank You,

Harsh.

To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/CAG_QrtL8q3ZYDCwxHb5eXy9mcJD3dgMR9%3Dijk-qSyZZ4sN0NbQ%40mail.gmail.com.

[Frank Yeh]

For --action=exp, please specify --export=nqa,iso

> To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/8C476E84-9E16-4D77-809D-7B166C3F5F38%40g.rit.edu.

[Harsh Prajapati]

Hello Frank,

Thanks a lot for your quick fixes, the registration seems to be working fine now and I am getting significantly better R2 values. Now, I want to apply this warping to two other QA maps, I could easily apply it one map by using --apply_warp. However DSI studio throws an error when I specify two filenames. I tried providing two filenames using the following ways:

1. --apply_warp=filename1,filename2

2. --apply_warp=filename1 --apply_warp2=filename2

3. --apply_warp=filename1 --apply_warp=filename2

None of this worked. 

What would be the correct syntax to apply warping to two separate files?

Thank You,

Harsh.  

To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/CAG_QrtL3MO7Njp0LaxCwc082a99sDuwvzbrppAgXGre-XFGfkg%40mail.gmail.com.

[Frank Yeh]

I modified DSI Studio to accept multiple files in --apply_warp. You
can now specify files separated by a comma

--apply_warp=file1.nii.gz,file2.nii.gz,file3,nii.gz

Please download today's version and see if it works for you.

Thanks for the suggestion!

Best,
Frank

> To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/CANyyq9HQ2XpmPxPuw0ZiAgxKcP-p%3DW9tpJwX6QrcEO8ah6z%2Bxw%40mail.gmail.com.

[Harsh Prajapati]

Hello Frank,

Thanks, this works. However, I have two concerns:

1. After registering to the MNI space, I get the range of QA values [-0.04,0.99]. How to handle/interpret the negative values?

2. When registering QA maps to MNI space using --action=reg, the transformed QA map (after registration) saved by default (without using --apply_warp) has range [-4.1,13.45]. However, it is only when I use --apply_warp on the original QA map I get range [-0.04,0.99]

 --action=reg --from=QA.nii.gz --from2=ISO.nii.gz --to=HCP1065.QA.nii.gz --to2=HCP1065.ISO.nii.gz        range of transformed QA - [-4.1,13.45]

 --action=reg --from=QA.nii.gz --from2=ISO.nii.gz --to=HCP1065.QA.nii.gz --to2=HCP1065.ISO.nii.gz --apply_warp=QA.nii.gz    range of transformed QA - [-0.04,0.99]

Thank You,

Harsh.  

You received this message because you are subscribed to a topic in the Google Groups "DSI Studio" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/dsi-studio/kytdlVQgmY8/unsubscribe.
To unsubscribe from this group and all its topics, send an email to dsi-studio+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/CAG_QrtKriDFwewfxYwK%2BZEuQ56uTU4EKCTLtKUWMOZKHW5hiHA%40mail.gmail.com.

[Frank Yeh]

Sorry about this bug. I fixed the code.
Could you give it a try to see if the issue is fixed?
If not, I may need the files you used to trace the problem.


Thank you very much for reporting this issue.

Best regards,
Frank

> To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/CANyyq9FsZts4bqKwnaSvvOrt-aJwSaU4EO%2B1g3FpHdN-WHWh9w%40mail.gmail.com.

[Harsh Prajapati]

Hello Frank,

I am getting a 'floating point exception: 8' error.  Below is the command and the log

./dsi_studio.app/Contents/MacOS/dsi_studio --action=reg --from=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz.nqa.nii.gz --from2=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz.iso.nii.gz --to=/Users/harshprajapati/Desktop/dsi_studio.app/Contents/MacOS/template/HCP1065.QA.nii.gz --to2=/Users/harshprajapati/Desktop/dsi_studio.app/Contents/MacOS/template/HCP1065.ISO.nii.gz

DSI Studio Jun 21 2021, Fang-Cheng Yeh

action=reg

from=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz.nqa.nii.gz

to=/Users/harshprajapati/Desktop/dsi_studio.app/Contents/MacOS/template/HCP1065.QA.nii.gz

from2=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz.iso.nii.gz

to2=/Users/harshprajapati/Desktop/dsi_studio.app/Contents/MacOS/template/HCP1065.ISO.nii.gz

output=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_01/EMBC21_base_Time_01/pair_6.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20110801101206_6_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz.nqa.nii.gz.wp.nii.gz

running linear registration.

reg_type=1

0.495546 0.0210879 -0.00403892 23.8633

-0.0119777 0.496583 0.0277261 27.0382

0.000862732 -0.0267802 0.451045 11.3623

correlation cofficient: 0.714411

reg_type=1

normalize_signal=1

normalizing signals

nonlinear registration using dual image modalities

resolution=2

smoothness=0.3

constraint=0.5

iteration=60

min_dimension=32

Floating point exception: 8

I downloaded the latest version of dsi studio for mac OS from the box link.

Thank You,

Harsh. 

To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/CAG_QrtJYhLNY_8BjL1PAf%3Dv6mxZTQCpcCYUP2xf1JpnCx_Qc7Q%40mail.gmail.com.

[Frank Yeh]

Sorry about the bug. I am fixing it now and will get back to you as
soon as possible.
Best regards,
Frank

> To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/CANyyq9HQ3%2B68VSeqEJrou6BtPGkm7S-gwuSH%2BOH%2BrzNNyZ27-g%40mail.gmail.com.

[Frank Yeh]

I updated DSI Studio to fix the bug.
Please download today's build and see if it works for you.

My mistake for not testing it thoroughly, and thank you for your patience.
Best regards,
Frank

[Harsh Prajapati]

Hello Frank,

Thank you for the quick bug fixes. It works now. The only issue is the range of QA values I get after registration is [-0.04,1.01]. How do you handle/interpret the negative values?

Thank You,

Harsh. 

To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/CAG_QrtKhGttYE1HJtzZe7WqV9eJChPPFCow-KyHvKDzrafag3A%40mail.gmail.com.

[Frank Yeh]

This is because of the cubic interpolation used in generate wrapped images. 

The interpolated values can be higher/lower than the maximum/minimum in the background noise.

You should be able to eliminate those by applying a mask.

or, I revise DSI Studio to allow users to specify the linear interpolation by --interpolation=0   

You may update DSI Studio to see if it works in your data.

Best regards,

Frank

To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/CANyyq9GbAeTaYEwASz-sqRfaLtk2bxK5z-CBAgozCyxmC7meCQ%40mail.gmail.com.

[Harsh Prajapati]

Hello Frank,

Thank you for the linear interpolation option, it works great for my data.

However, the latest version of dsi studio for MacOs - JUL 2nd build is unable to export FA values. Below is the command and the associated error:

Command:

./dsi_studio.app/Contents/MacOS/dsi_studio --action=exp --source=/Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_02/EMBC21_base_Time_01/pair_5.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20111115155840_5_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz --export=fa

Error:

ERROR: cannot find fa in /Users/harshprajapati/Desktop/Brain_Imaging/EMBC21_reg/Subject_001/Time_02/EMBC21_base_Time_01/pair_5.ep2d_diff_DTI_60DWI_10b0_1avg_ep2d_diff_DTI_60DWI_10b0_1avg_20111115155840_5_b0corrected.nii.gz.src.gz.gqi.1.25.fib.gz

I also tried using --export=fa0, but that does not work either.

Thank You,

Harsh. 

To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/CAG_QrtJfdM5WYZZUa_n08tgk0xUzC5nVFE%3D8E-2%3DiT1e_PY8UQ%40mail.gmail.com.

[Frank Yeh]

--export=dti_fa

Frank

To view this discussion on the web visit https://groups.google.com/d/msgid/dsi-studio/CANyyq9Gi8e_jPiAy-hM937mXiu13Sn%3Dc-4f_HF%2BeF_SBE9kz_Q%40mail.gmail.com.