Difference between revisions of "2009 Winter Project Week:LesionSegmentationEfficiency"

From NAMIC Wiki
Jump to: navigation, search
 
(5 intermediate revisions by 2 users not shown)
Line 1: Line 1:
 +
{|
 +
|[[Image:NAMIC-SLC.jpg|thumb|320px|Return to [[2009_Winter_Project_Week|Project Week Main Page]] ]]
 +
|}
 +
 +
 +
__NOTOC__
 +
 +
 
===Key Investigators===
 
===Key Investigators===
 
Mark Scully, MIND  
 
Mark Scully, MIND  
  
Jeremy Bockholdt, MIND
+
Jeremy Bockholt, MIND
  
 
Steve Piper
 
Steve Piper
 +
 +
Marco Ruiz
 +
 +
Luis Ibanez
  
 
<div style="margin: 20px;">
 
<div style="margin: 20px;">
Line 11: Line 23:
  
 
<h1>Objective</h1>
 
<h1>Objective</h1>
Improve the running time and memory use of the new lesion segmentation method.
+
Improve the running time and memory use of the new lesion segmentation method as the current module takes 90-120 minutes and up to 2 Gb of RAM.
  
 
</div>
 
</div>
Line 21: Line 33:
 
Make the custom pieces that can't be replaced multi-thread capable.
 
Make the custom pieces that can't be replaced multi-thread capable.
 
Locate inefficient memory use and improve it.
 
Locate inefficient memory use and improve it.
 +
 +
See what if anything could be made to work in GWE (Marcus)?
 
</div>
 
</div>
  
Line 26: Line 40:
  
 
<h1>Progress</h1>
 
<h1>Progress</h1>
 
+
# Luiz provided an itk statistics filter to replace current custom, single threaded, un-optimized code.
 +
# Final piece on single threaded code in process of being replaced by a custom multi-threaded ITK filter.
 +
# Prediction runtime reduced from between 1.5 - 2 hours to less than 9 minutes on a quad core machine.
 +
# Prediction Memory footprint reduced from 2.5GB to 1.1GB
 +
# With Marco's help our pipeline is now running on Grid Wizard
  
 
</div>
 
</div>

Latest revision as of 22:55, 8 January 2009

Home < 2009 Winter Project Week:LesionSegmentationEfficiency



Key Investigators

Mark Scully, MIND

Jeremy Bockholt, MIND

Steve Piper

Marco Ruiz

Luis Ibanez

Objective

Improve the running time and memory use of the new lesion segmentation method as the current module takes 90-120 minutes and up to 2 Gb of RAM.

Approach, Plan

Review code to determine custom pieces that can be replaced with existing filters. Make the custom pieces that can't be replaced multi-thread capable. Locate inefficient memory use and improve it.

See what if anything could be made to work in GWE (Marcus)?

Progress

  1. Luiz provided an itk statistics filter to replace current custom, single threaded, un-optimized code.
  2. Final piece on single threaded code in process of being replaced by a custom multi-threaded ITK filter.
  3. Prediction runtime reduced from between 1.5 - 2 hours to less than 9 minutes on a quad core machine.
  4. Prediction Memory footprint reduced from 2.5GB to 1.1GB
  5. With Marco's help our pipeline is now running on Grid Wizard


References