Difference between revisions of "2008 Winter Project Week GroupwiseReg"

From NAMIC Wiki
Jump to: navigation, search
 
Line 10: Line 10:
 
===Key Investigators===
 
===Key Investigators===
 
* MIT: Serdar Balci, Polina Golland
 
* MIT: Serdar Balci, Polina Golland
* Kitware: Luis Ibanez
+
* Kitware: Brad Davis
 
+
* SCI: Casey Goodlett
  
 
<div style="margin: 20px;">
 
<div style="margin: 20px;">
Line 19: Line 19:
 
<h1>Objective</h1>
 
<h1>Objective</h1>
  
The goal of this project is to develop an efficient B-Splines based registration code.
+
The goal of this project is to expand the NAMIC Software Kit with the addition of an image-based atlas
We explore the space of different optimization schemes e.g. caching, sparsity and memory-speed trade-off's.
+
building algorithm that was developed by Polina Golland’s lab at MIT. The basic functionality of this software will
 +
be demonstrated to the NAMIC community at the January 2008 NAMIC all-hands meeting. A secondary
 +
goal is to begin the process of gathering usability and interface requests from targeted clinical collaborators.
 +
It is intended that users’ requests will be incorporated in to a future project aimed at tailoring this software
 +
for widespread NAMIC use.
 +
 
 +
See our [[Projects:GroupwiseRegistration| Algorithm Project Page]] for more details.
  
 
</div>
 
</div>
Line 28: Line 34:
 
<h1>Approach, Plan </h1>
 
<h1>Approach, Plan </h1>
  
We run benchmark experiments in the special case of itkMattesMutualInformation implementation.
+
The registration algorithms are already available as C++ sourcecode. We plan to integrate them into the NAMIC community by:
The current implementation in ITK uses caching of B-Spline coefficient in order to speed-up
 
the registration. Also the sparsity of B-Spline coefficient updates are taken into account, yielding
 
and efficient implementation.
 
  
The drawback of caching is the use of high amounts of memory.  
+
*Adapt sourcecode to utilize the command-line parsing mechanism that was designed for Slicer3 external modules.
We compare the performance of caching vs. not-caching(computing the coefficients on the fly)
+
*Create a simple Slicer3 graphical user interface (external module) that allows a user to generate an atlas from their data.  
by deactivating caching code in itkMattesMutualInformation .
+
*User Feedback. During the January 2008 NAMIC all-hands meeting we will solicit feedback on this project from potential collaborators.
  
 
</div>
 
</div>
Line 46: Line 49:
 
====Jan 2008 Project Week====
 
====Jan 2008 Project Week====
  
 +
</div>
  
We run an experiment on a machine
+
<br style="clear: both;" />
with 8GB of memory, 4 core CPU
 
running gcc compiler on a linux system.
 
 
 
We used a B-Spline with mesh size 32x32x16
 
with an image of size 256x256x124 and used
 
Mattes mutual information as the registration algorithm.
 
 
 
Run-times:
 
* 21 seconds with caching (2.4 GB of memory)
 
* 52 seconds w/o caching (800MB of memory)
 
 
 
Overall in this experiment, caching improved the performance by 2.5 while using 3 times more memory.
 
We are in the process of exploring the effect of number of samples, image size and compiler on
 
registration time.
 
  
 
</div>
 
</div>
  
<br style="clear: both;" />
 
  
</div>
+
===References===
 +
* S.K. Balci, P. Golland, M.E. Shenton, W.M. Wells III. Free-Form B-spline Deformation Model for Groupwise Registration. In Proceedings of MICCAI 2007 Statistical Registration Workshop: Pair-wise and Group-wise Alignment and Atlas Formation, 23-30, 2007.

Latest revision as of 15:17, 11 January 2008

Home < 2008 Winter Project Week GroupwiseReg
Figure 1: On the left is shown a stack of images and a sample pixel stack around a cortical region. On the left is shown the Gaussian(blue) fittet to a real sample from the dataset we used along with the non-parametric density estimate(red). Note that the distribution is bi-modal because of white matter-gray matter transaction.


Key Investigators

  • MIT: Serdar Balci, Polina Golland
  • Kitware: Brad Davis
  • SCI: Casey Goodlett

Objective

The goal of this project is to expand the NAMIC Software Kit with the addition of an image-based atlas building algorithm that was developed by Polina Golland’s lab at MIT. The basic functionality of this software will be demonstrated to the NAMIC community at the January 2008 NAMIC all-hands meeting. A secondary goal is to begin the process of gathering usability and interface requests from targeted clinical collaborators. It is intended that users’ requests will be incorporated in to a future project aimed at tailoring this software for widespread NAMIC use.

See our Algorithm Project Page for more details.

Approach, Plan

The registration algorithms are already available as C++ sourcecode. We plan to integrate them into the NAMIC community by:

  • Adapt sourcecode to utilize the command-line parsing mechanism that was designed for Slicer3 external modules.
  • Create a simple Slicer3 graphical user interface (external module) that allows a user to generate an atlas from their data.
  • User Feedback. During the January 2008 NAMIC all-hands meeting we will solicit feedback on this project from potential collaborators.

Progress


Jan 2008 Project Week



References

  • S.K. Balci, P. Golland, M.E. Shenton, W.M. Wells III. Free-Form B-spline Deformation Model for Groupwise Registration. In Proceedings of MICCAI 2007 Statistical Registration Workshop: Pair-wise and Group-wise Alignment and Atlas Formation, 23-30, 2007.