Difference between revisions of "2009 Summer Project Week Registration for RT"

From NAMIC Wiki
Jump to: navigation, search
 
(5 intermediate revisions by the same user not shown)
Line 1: Line 1:
 
 
__NOTOC__
 
__NOTOC__
 
<gallery>
 
<gallery>
Image:PW2009-v3.png|[[2009_Summer_Project_Week|Project Week Main Page]]
+
Image:PW2009-v3.png|[[2009_Summer_Project_Week#Projects|Project Week Main Page]]
Image:genuFAp.jpg|Scatter plot of the original FA data through the genu of the corpus callosum of a normal brain.
 
Image:genuFA.jpg|Regression of FA data; solid line represents the mean and dotted lines the standard deviation.
 
 
</gallery>
 
</gallery>
  
Line 11: Line 8:
 
==Key Investigators==
 
==Key Investigators==
 
* BWH: Sandy Wells, Tina Kapur
 
* BWH: Sandy Wells, Tina Kapur
* U Michigan: James Balter
+
* U Michigan: James Balter, Dan McShan
 
* MGH: Lilla Zollei, Greg Sharp
 
* MGH: Lilla Zollei, Greg Sharp
 +
* NVIDIA: Joe Stam
 +
*Mayo: Dan Blezek
 +
*Kitware: Sebastien Barre
  
 
<div style="margin: 20px;">
 
<div style="margin: 20px;">
Line 32: Line 32:
  
  
Our plan for the project week is to first try out <bar>,...
+
DRR creation is a compute+memory intensive task of our 2d/3d registration algorithms and we plan to accelerate that using CUDA.
 
 
 
</div>
 
</div>
  
Line 39: Line 38:
  
 
<h3>Progress</h3>
 
<h3>Progress</h3>
To be filled in later.
+
We achieved the following:
 
+
*started cuda acceleration of the drr creation routine from plastimatch (greg sharp)
 +
*started using cmake for 2d/3d registration code base from mvis (sandy wells)
 +
*obtained phantom data (CT and projections) from U Michigan
  
 
</div>
 
</div>

Latest revision as of 12:55, 26 June 2009

Home < 2009 Summer Project Week Registration for RT


Key Investigators

  • BWH: Sandy Wells, Tina Kapur
  • U Michigan: James Balter, Dan McShan
  • MGH: Lilla Zollei, Greg Sharp
  • NVIDIA: Joe Stam
  • Mayo: Dan Blezek
  • Kitware: Sebastien Barre

Objective

We will be taking an inventory of 2d/3d registration algorithms available between our groups, and accelerating at least one using CUDA for NVIDIA hardware.




Approach, Plan


DRR creation is a compute+memory intensive task of our 2d/3d registration algorithms and we plan to accelerate that using CUDA.

Progress

We achieved the following:

  • started cuda acceleration of the drr creation routine from plastimatch (greg sharp)
  • started using cmake for 2d/3d registration code base from mvis (sandy wells)
  • obtained phantom data (CT and projections) from U Michigan

References