Difference between revisions of "Summer2009:Using CUDA for stochastic tractography"

From NAMIC Wiki
Jump to: navigation, search
 
(One intermediate revision by the same user not shown)
Line 28: Line 28:
 
<h3>Progress</h3>
 
<h3>Progress</h3>
 
* Cuda was integrated through PyCuda with support for numpy arrays syntax
 
* Cuda was integrated through PyCuda with support for numpy arrays syntax
* direct call to driver api tested
+
* Check compilation&instalation under Linux&Windows
* driver kernels tested (simple matrix operations)
+
* Implemented part of the tractography algorithm in kernel
 +
* Tested direct call to driver api - ok
 +
* Tested driver kernels (simple matrix operations) - ok
 +
* Still need to complete the whole tractography algorithm in kernel mode
 
   
 
   
  

Latest revision as of 22:07, 26 June 2009

Home < Summer2009:Using CUDA for stochastic tractography


Key Investigators

  • BWH: Julien de Siebenthal, Sylvain Bouix

Objective

Stochastic tractography does not provide interactive visualization so far due to its intensive computational needs.

Approach, Plan

Idea would be to visualize online paths generated in a point of interest like a fiducial. This approach would be based on the online visualization of streamline tractography done by moving a fiducial interactively.

During the summer week, we will continue our work to develop a concrete solution in investigating acceleration means based mainly on CUDA.

Progress

  • Cuda was integrated through PyCuda with support for numpy arrays syntax
  • Check compilation&instalation under Linux&Windows
  • Implemented part of the tractography algorithm in kernel
  • Tested direct call to driver api - ok
  • Tested driver kernels (simple matrix operations) - ok
  • Still need to complete the whole tractography algorithm in kernel mode



References