failure of the autotrack feature

[drobwolf]

Hi, thanks for your fantastic work! I'm running into an issue with some reconstructions using the autotrack feature in the CLI in which it generates only a file ending in "no_result.txt", but no error messages are generated in stderr and everything in stdout looks good...i'm not sure where to begin debugging. I saw from your autotrack code that this can happen either when a file by that name is already present and overwrite is set to off, or when there are "too low yield" of fibers due to quality issues...but I don't believe either of those is occurring in my case.

[frankyeh]

Have you done a quality check on your data? http://dsi-studio.labsolver.org/Manual/Parse-DICOM#TOC-Batch-Quality-Control-for-SRC-files

Also, I would suggest updating DSI Studio again and see if that works.

If still doesn't work, please send me an SRC file, and I will help check the problem for you.

Best regards, Frank

On Thu, Jul 22, 2021 at 4:31 PM drobwolf @.***> wrote:

Hi, thanks for your fantastic work! I'm running into an issue with some reconstructions using the autotrack feature in the CLI in which it generates only a file ending in "no_result.txt", but no error messages are generated in stderr and everything in stdout looks good...i'm not sure where to begin debugging. I saw from your autotrack code that this can happen either when a file by that name is already present and overwrite is set to off, or when there are "too low yield" of fibers due to quality issues...but I don't believe either of those is occurring in my case.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/frankyeh/DSI-Studio/issues/67, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACDI4MLWJ4J6FJXT7SNCJDTZB53DANCNFSM5A2XODYA .

[drobwolf]

Thanks for the quick reply! When I qc using the CLI, I get the following:

Fiber coherence index: 6251.8
Fiber incoherent index: 37335.6

Just the two numbers for each file. I'm wondering whether this might be due to the fact that my "source" files are all "fib.gz"?

[frankyeh]

Yes, please assign SRC files and see if it works. Frank

On Thu, Jul 22, 2021 at 7:16 PM drobwolf @.***> wrote:

Thanks for the quick reply! When I qc using the CLI, I get the following:

Fiber coherence index: 6251.8 Fiber incoherent index: 37335.6

Just the two numbers for each file. I'm wondering whether this might be due to the fact that my "source" files are all "fib.gz"?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/frankyeh/DSI-Studio/issues/67#issuecomment-885296867, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACDI4LXARJDQQTCFYXX6PDTZCRGBANCNFSM5A2XODYA .

[drobwolf]

so i dont seem to have src files for these subjects. their fib.gz files were generated prior to my joining the lab using the qsiprep pipeline. id be happy to send a fib.gz file but github wont seem to attach it (it is around 400 mb), so ive included a google drive link https://drive.google.com/file/d/1JQ4OdfL4wjLYSDi1I7GCzfX8skFp1W6H/view?usp=drivesdk

[frankyeh]

I checked the data in detail, and there is bad news. The b-table has a problem.

Please check the cingulum in detail (highlighted) and you will see the fiber orientation is slightly rotated downward. A likely cause is that the images were resampled to align AC-PC, but the b-table was not rotated accordingly. The preprocessing step has to be corrected.

The only solution is to start from the raw data and re-process them from the bottom-up (even the SRC files need to be re-created).

I am sorry about this, but tracking will not work out unless a correct table was used.

Best regards, Frank

[image: image.png]

On Thu, Jul 22, 2021 at 9:58 PM drobwolf @.***> wrote:

so i dont seem to have src files for these subjects. their fib.gz files were generated prior to my joining the lab using the qsiprep pipeline. id be happy to send a fib.gz file but github wont seem to attach it (it is around 400 mb), so ive included a google drive link

https://drive.google.com/file/d/1JQ4OdfL4wjLYSDi1I7GCzfX8skFp1W6H/view?usp=drivesdk

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/frankyeh/DSI-Studio/issues/67#issuecomment-885348293, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACDI4IPBRIL75O2TVAVZIDTZDED3ANCNFSM5A2XODYA .

[drobwolf]

Thanks again for looking into this so quickly. The image you referenced in your reply somehow didn't come through for me, could you try re-sending/re-attaching it?

[frankyeh]

[mattcieslak]

@drobwolf did you create these with the dsistudio_gqi pipeline from qsiprep? If not and you tried to use dsi studio's most recent release directly DSI Studio will interpret the bvecs incorrectly. You'll need to use a flag that tells DSI Studio to interpret bvecs how it used to circa early 2020 (I forget flag offhand)

[frankyeh]

It is --flip_by=1

On Fri, Jul 30, 2021 at 1:37 PM Matt Cieslak @.***> wrote:

@drobwolf https://github.com/drobwolf did you create these with the dsistudio_gqi pipeline from qsiprep? If not and you tried to use dsi studio's most recent release directly DSI Studio will interpret the bvecs incorrectly. You'll need to use a flag that tells DSI Studio to interpret bvecs how it used to circa early 2020 (I forget flag offhand)

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/frankyeh/DSI-Studio/issues/67#issuecomment-890048905, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACDI4PCL64NBVA4HCTWEMDT2LPOZANCNFSM5A2XODYA .

[drobwolf]

Thank you both for your very helpful comments! We re-ran qsiprep using v 0.14. I then subsequently ran dsistudio within a singularity sif (pulled since may 2021) with the following syntax:

bvec=$(echo .bvec) && bval=$(echo .bval) && source=$(echo *dwi.nii.gz) && singularity exec ~/software/dsistudio_latest.sif dsi_studio --action=atk --source=$source --bval=$bval --bvec=$bvec --flip_by=1

However I am still failing to track some files, resulting in stat files that end in '*no_result.txt'.

Below I have included a link to a Dropbox folder with the files needed to reproduce this problem. I'm wondering whether you both could take a look when you have the time to do so?

Best, Doug wolfe

https://www.dropbox.com/sh/xcbz4jlydo32duw/AAB5i-fuzRqP2itlK4uelVSIa?dl=0

[frankyeh]

Hi Doug,

You may check this out:

http://dsi-studio.labsolver.org/Manual/automatic-fiber-tracking#TOC-Troubleshooting

If "no result" happens quite often (e.g. > 50%), then there is a

quality issue (e.g., b-table flipped) that needs to be corrected.

If it happens occasionally (e.g. < 5%) for some smaller pathways (e.g.,

right AF, SLF I...etc.), then it is likely that your data acquisition is not good enough (check if data have any of the conditions listed in the documentation).

Best, Frank

On Tue, Sep 21, 2021 at 4:42 PM drobwolf @.***> wrote:

Thank you both for your very helpful comments! We re-ran qsiprep using v 0.14. I then subsequently ran dsistudio within a singularity sif (pulled since may 2021) with the following syntax:

bvec=$(echo .bvec) && bval=$(echo .bval) && source=$(echo *dwi.nii.gz) && singularity exec ~/software/dsistudio_latest.sif dsi_studio --action=atk --source=$source --bval=$bval --bvec=$bvec --flip_by=1

However I am still failing to track some files, resulting in stat files that end in '*no_result.txt'.

Below I have included a link to a Dropbox folder with the files needed to reproduce this problem. I'm wondering whether you both could take a look when you have the time to do so?

Best, Doug wolfe

https://www.dropbox.com/sh/xcbz4jlydo32duw/AAB5i-fuzRqP2itlK4uelVSIa?dl=0

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/frankyeh/DSI-Studio/issues/67#issuecomment-924368008, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACDI4OU7VOIRX6V5TMZVWLUDDUZXANCNFSM5A2XODYA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

[drobwolf]

Wonderful, thanks for your quick replies as always!

Sent from my iPhone

On Sep 21, 2021, at 6:57 PM, Fang-Cheng (Frank) Yeh @.***> wrote:

 Hi Doug,

You may check this out: http://dsi-studio.labsolver.org/Manual/automatic-fiber-tracking#TOC-Troubleshooting

If "no result" happens quite often (e.g. > 50%), then there is a quality issue (e.g., b-table flipped) that needs to be corrected.

If it happens occasionally (e.g. < 5%) for some smaller pathways (e.g., right AF, SLF I...etc.), then it is likely that your data acquisition is not good enough (check if data have any of the conditions listed in the documentation).

Best, Frank

On Tue, Sep 21, 2021 at 4:42 PM drobwolf @.***> wrote:

Thank you both for your very helpful comments! We re-ran qsiprep using v 0.14. I then subsequently ran dsistudio within a singularity sif (pulled since may 2021) with the following syntax:

bvec=$(echo .bvec) && bval=$(echo .bval) && source=$(echo *dwi.nii.gz) && singularity exec ~/software/dsistudio_latest.sif dsi_studio --action=atk --source=$source --bval=$bval --bvec=$bvec --flip_by=1

However I am still failing to track some files, resulting in stat files that end in '*no_result.txt'.

Below I have included a link to a Dropbox folder with the files needed to reproduce this problem. I'm wondering whether you both could take a look when you have the time to do so?

Best, Doug wolfe

https://www.dropbox.com/sh/xcbz4jlydo32duw/AAB5i-fuzRqP2itlK4uelVSIa?dl=0

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/frankyeh/DSI-Studio/issues/67#issuecomment-924368008, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACDI4OU7VOIRX6V5TMZVWLUDDUZXANCNFSM5A2XODYA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

[frankyeh]

Just revised the document a few seconds ago. Hope this helps.

On Tue, Sep 21, 2021 at 7:02 PM drobwolf @.***> wrote:

Wonderful, thanks for your quick replies as always!

Sent from my iPhone

On Sep 21, 2021, at 6:57 PM, Fang-Cheng (Frank) Yeh @.***> wrote:

 Hi Doug,

You may check this out:

http://dsi-studio.labsolver.org/Manual/automatic-fiber-tracking#TOC-Troubleshooting

If "no result" happens quite often (e.g. > 50%), then there is a quality issue (e.g., b-table flipped) that needs to be corrected.

If it happens occasionally (e.g. < 5%) for some smaller pathways (e.g., right AF, SLF I...etc.), then it is likely that your data acquisition is not good enough (check if data have any of the conditions listed in the documentation).

Best, Frank

On Tue, Sep 21, 2021 at 4:42 PM drobwolf @.***> wrote:

Thank you both for your very helpful comments! We re-ran qsiprep using v 0.14. I then subsequently ran dsistudio within a singularity sif (pulled since may 2021) with the following syntax:

bvec=$(echo .bvec) && bval=$(echo .bval) && source=$(echo *dwi.nii.gz) && singularity exec ~/software/dsistudio_latest.sif dsi_studio --action=atk --source=$source --bval=$bval --bvec=$bvec --flip_by=1

However I am still failing to track some files, resulting in stat files that end in '*no_result.txt'.

Below I have included a link to a Dropbox folder with the files needed to reproduce this problem. I'm wondering whether you both could take a look when you have the time to do so?

Best, Doug wolfe

https://www.dropbox.com/sh/xcbz4jlydo32duw/AAB5i-fuzRqP2itlK4uelVSIa?dl=0

— You are receiving this because you commented. Reply to this email directly, view it on GitHub < https://github.com/frankyeh/DSI-Studio/issues/67#issuecomment-924368008>, or unsubscribe < https://github.com/notifications/unsubscribe-auth/AACDI4OU7VOIRX6V5TMZVWLUDDUZXANCNFSM5A2XODYA

. Triage notifications on the go with GitHub Mobile for iOS < https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675

or Android < https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub .

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/frankyeh/DSI-Studio/issues/67#issuecomment-924450067, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACDI4IYZRDZJHUPJYPL5OLUDEFIPANCNFSM5A2XODYA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

[drobwolf]

I believe I've fixed the b table issue based on the attached fib.gz, but I'm still failing with autotrack atk on multiple tracks. I'm a bit stuck and I'm wondering whether you could confirm whether this fib file looks alright? If so what else might be causing the failure of atk?

https://www.dropbox.com/s/tcghmjw1sutg3xk/sub-NDARKN305AUG_ses-01_run-001_space-T1w_desc-preproc_space-T1w_fslstd_dwi.nii.gz.gqi.1.25.fib.gz?dl=0

[frankyeh]

Have you checked the data using the instruction here? http://dsi-studio.labsolver.org/Manual/automatic-fiber-tracking#TOC-Troubleshooting

If this still doesn't help to find out the cause, please let me know. Frank

On Thu, Oct 21, 2021 at 11:59 AM drobwolf @.***> wrote:

I believe I've fixed the b table issue based on the attached fib.gz, but I'm still failing with autotrack atk on multiple tracks. I'm a bit stuck and I'm wondering whether you could confirm whether this fib file looks alright? If so what else might be causing the failure of atk?

https://www.dropbox.com/s/tcghmjw1sutg3xk/sub-NDARKN305AUG_ses-01_run-001_space-T1w_desc-preproc_space-T1w_fslstd_dwi.nii.gz.gqi.1.25.fib.gz?dl=0

— You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub https://github.com/frankyeh/DSI-Studio/issues/67#issuecomment-948753741, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACDI4IRDLSVF2VM6VLCIVDUIA2EZANCNFSM5A2XODYA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.