[Mrtrix-discussion] Fwd: mrtrix+nipype workflow

[Arnaud Guidon]

For the sake of correctness and in case someone else was looking into this, the actual parameter to set is:

probCSDstreamtrack.inputs.desired_number_of_tracks = 10000



Begin forwarded message:

> From: Arnaud Guidon <ag114 at duke.edu>
> Subject: Re: mrtrix+nipype workflow
> Date: August 3, 2012 6:05:11 PM EDT
> To: mrtrix-discussion at www.nitrc.org
> 
> Well, with regard to my previous post, things start looking a lot better after adjusting the parameters.  The default value of 1000 tracks just doesn't seem to be generous enough for the FSL data.  Changing the desired number of tracks in the script (http://bit.ly/MVbZvp) helps greatly:
> 
> probCSDstreamtrack.inputs.number = 10000
> 
> Arnaud
> 
> 
> Begin forwarded message:
> 
>> From: Arnaud Guidon <ag114 at duke.edu>
>> Subject: mrtrix+nipype workflow
>> Date: August 3, 2012 4:34:48 PM EDT
>> To: mrtrix-discussion at www.nitrc.org
>> 
>> Hi all,
>> 
>> I am currently evaluating the mrtrix+nipype workflow (http://bit.ly/MVbZvp) for a study where I would need to consider multifiber tracking based on CSD.
>> 
>> I followed the instructions and successfully ran the mrtrix pipeline on the FSL course data.  However, the results seem to drastically underperform conventional streamline tractography.  I am aware that there is a lot of refactoring going on in some of the dependencies used to run this pipeline (including dipy 0.6) which potentially could explain these puzzling results (see attached: http://bit.ly/QnZPKP)
>> 
>> Did anyone get better results after running this workflow? It would be great to have an idea of what the results should really look like.  
>> 
>> I'd appreciate your thoughts on this.
>> 
>> Thanks,
>> 
>> Arnaud
>> 
> 

Arnaud Guidon, PhD Candidate
Biomedical Engineering
Duke University
ag114 at duke.edu
919-428-1801

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.nitrc.org/pipermail/mrtrix-discussion/attachments/20120803/b739d8a1/attachment.html