Mrtrix3

[Jason Yeatman]

Hey AFQ users,

We have been working on rewriting the AFQ-mrtrix interface to work with mrtrix3 and to handle multi shell data. There is an open pull request (see mrtrix3) branch. We welcome contributions from anyone that wants to test this on their data.

Best,

Jason

--
Sent from a mobile device. Please excuse typos

[Michael Zeineh]

Hi Jason,

So funny, I've just been working on this too! I checked here recently but not recently enough, I should have rechecked here first! If it helps, here are the minor code modifications I have made.

I've been having the following problem. After trying my modified code, I get the following errors. Any ideas? The tracks look fine when viewed with mrview.

dtiSplitInterhemisphericFibers: Splitting every fiber below Z=-10
Error using horzcat
Dimensions of matrices being concatenated are not consistent.

Error in dtiIntersectFibersWithRoi (line 140)
    fc = horzcat(fg.fibers{:})';

Error in dtiSplitInterhemisphericFibers (line 67)
    [fgToChop,contentiousFibers, keep] = dtiIntersectFibersWithRoi([], {'and'}, [], roi, fg);


Error in AFQ_SegmentFiberGroups (line 205)
    fg      = dtiSplitInterhemisphericFibers(fg, dt, -10);


Error in AFQ_run (line 169)
        fg_classified = AFQ_SegmentFiberGroups(dtFile, fg, [], [],[], antsInvWarp);

[Jason Yeatman]

Hmm, I don't know what would cause that error.

--
You received this message because you are subscribed to the Google Groups "Automated Fiber Quantification (AFQ)" group.
To unsubscribe from this group and stop receiving emails from it, send an email to automated-fiber-quantification-afq+unsub...@googlegroups.com.
To post to this group, send email to automated-fiber-quantification-afq@googlegroups.com.
Visit this group at https://groups.google.com/group/automated-fiber-quantification-afq.
For more options, visit https://groups.google.com/d/optout.

[Michael Zeineh]

This ended up fixing itself using your new code. I do have some minor modifications I made to your code, what is the best way to inform?

[Jason Yeatman]

Can you submit a pull request over github? That is the best way for us to go through, check and accept suggestions on the code

https://help.github.com/articles/creating-a-pull-request/

Thanks

Jason

To unsubscribe from this group and stop receiving emails from it, send an email to automated-fiber-quantification-afq+unsubscribe@googlegroups.com.
To post to this group, send email to automated-fiber-quantification-a...@googlegroups.com.

Visit this group at https://groups.google.com/group/automated-fiber-quantification-afq.
For more options, visit https://groups.google.com/d/optout.

[Luie Wang]

Hi all, 

Sorry to bother you from a old post.... I encountered the exact same error when running AFQ_run. I have created the whole brain tractography using multi-shell from MRtrix and have converted & named them properly in the fibers folder. When running AFQ_run, the following warning message keeps showing up: (I have 1 M streamlines in my wholebrain tractography and the warning persists as they handle fibers from beginning to the end) 

Warning: Fiber 999997: Found aberrant fiber points, but coulnd decide which segment

to remove. 

> In dtiCleanFibers (line 64)

  In dtiSplitInterhemisphericFibers (line 52)

  In AFQ_SegmentFiberGroups (line 221)

  In AFQ_run (line 169) 

In addition, I received the following error message: 

dtiCleanFibers: Keeping 800245 out of 1000000 fibers.

dtiCleanFibers: Clipped 307297 aberrant fiber segments.

dtiSplitInterhemisphericFibers: Splitting every fiber below Z=-10 

Error using horzcat

Dimensions of arrays being concatenated are not consistent.

Error in dtiIntersectFibersWithRoi (line 140)

    fc = horzcat(fg.fibers{:})';

Error in dtiSplitInterhemisphericFibers (line 67)

    [fgToChop,contentiousFibers, keep] = dtiIntersectFibersWithRoi([], {'and'}, [],

    roi, fg);

Error in AFQ_SegmentFiberGroups (line 221)

    fg      = dtiSplitInterhemisphericFibers(fg, dt, -10);

Error in AFQ_run (line 169)

        fg_classified = AFQ_SegmentFiberGroups(dtFile, fg, [], [],[], antsInvWarp);

 

I analyze the tracks in both MRtrix and AFQ and they look fine. They also work perfectly when I try the single subject analysis but for somehow just didn't work with AFQ_run. How should I solve this problem? Any inputs/suggestion is greatly appreciated! Sorry for my frequent post..(just started exploring DTI tractography as a beginner recently) and thank you so much for your time and help!

Best,

Luie

To unsubscribe from this group and stop receiving emails from it, send an email to automated-fiber-quantification-afq+unsub...@googlegroups.com.
To post to this group, send email to automated-fiber-quantification-afq@googlegroups.com.
Visit this group at https://groups.google.com/group/automated-fiber-quantification-afq.
For more options, visit https://groups.google.com/d/optout.