[Insight-developers] Comments please: Prototype DTI fiber trackingin ITK

[Lorensen, William E (Research)]

I suggest we hold off committing this new code until after the release. We prefer not to change API's of releases and we should spend some time evaluating these new classes.

Bill


-----Original Message-----
From: insight-developers-bounces+lorensen=crd.ge.com at itk.org
[mailto:insight-developers-bounces+lorensen=crd.ge.com at itk.org]On Behalf
Of Torsten Rohlfing
Sent: Monday, August 08, 2005 3:46 PM
To: Karthik Krishnan
Cc: insight-developers at itk.org
Subject: Re: [Insight-developers] Comments please: Prototype DTI fiber
trackingin ITK


Hi Karthik --

Thanks for your feedback.

Karthik Krishnan wrote:

>>
>> The tracking filter inherits from itk::SpatialObjectSource (also 
>> included in aforementioned archive) and generates an 
>> itk::GroupSpatialObject as its output. The group object contains all 
>> fibers as its children, where each fiber is represented by an 
>> itk::DTITubeSpatialObject object.
>>
> I like the idea of having a SpatialObjectSource. For consistency, I 
> think the SpaitialObjectReader should also derive from this. With the 
> current layout, of the class hierarchy, it doesn't really fit in that 
> well. Julien will have more to say on that.

To tell you the truth, I was quite surprised that there was no 
SpatialObjectSource already.

>> I also include a first, simple test in itkFiberTrackingTest.cxx, 
>> which basically constructs a 5x5x5 tensor volume with a single 
>> straight "fiber" in z-direction in the center of the volume. Note 
>> that since the tensors are all exactly vertical, you'll basically get 
>> five identical fibers as output. I am not saying it's particularly 
>> creative test case.
>
>
> If you are ok with it, we could commit this to ITK.

Fundamentally, I am okay with it. However, I am not entirely happy with 
the iterator design yet. At the very least I would like create a traits 
class for the code that takes a tensor and computes the next line 
segment direction. This way, we could easily plug different such 
algorithms into the fiber iterator. For example, a Monte Carlo-type 
tracking would then simply be realized by pluging in a different such 
traits class.

This is a matter of a few hours of thinking and re-arranging things on 
my part, so if you don't mind, I'd suggest you give me another day and 
then I'll post an updated version that is more elegantly designed.

-- 
Torsten Rohlfing, PhD          SRI International, Neuroscience Program
 Research Scientist             333 Ravenswood Ave, Menlo Park, CA 94025
  Phone: ++1 (650) 859-3379      Fax: ++1 (650) 859-2743
   torsten at synapse.sri.com        http://www.stanford.edu/~rohlfing/

     "Though this be madness, yet there is a method in't"

_______________________________________________
Insight-developers mailing list
Insight-developers at itk.org
http://www.itk.org/mailman/listinfo/insight-developers