Difference between revisions of "2010 Summer Project Week Fiducial Deformable Registration"

From NAMIC Wiki
Jump to: navigation, search
 
(9 intermediate revisions by the same user not shown)
Line 2: Line 2:
 
<gallery>
 
<gallery>
 
Image:PW-MIT2010.png|[[2010_Summer_Project_Week#Projects|Projects List]]
 
Image:PW-MIT2010.png|[[2010_Summer_Project_Week#Projects|Projects List]]
−
Image:genuFAp.jpg|Scatter plot of the original FA data through the genu of the corpus callosum of a normal brain.
+
Image:fig-registr-wiki.jpg|Top: Reference and test synthetic images. Bottom: Results for registration without landmark information and using one landmark.
−
Image:genuFA.jpg|Regression of FA data; solid line represents the mean and dotted lines the standard deviation.
+
 
 +
Image:spring.png|Registration error vs number of landmarks.
 
</gallery>
 
</gallery>
  
Line 9: Line 10:
  
 
==Key Investigators==
 
==Key Investigators==
−
* UNC: Isabelle Corouge, Casey Goodlett, Guido Gerig
+
* MGH: Nadya Shusharina, Greg Sharp
−
* Utah: Tom Fletcher, Ross Whitaker
 
  
 
<div style="margin: 20px;">
 
<div style="margin: 20px;">
Line 29: Line 29:
 
<h3>Approach, Plan</h3>
 
<h3>Approach, Plan</h3>
  
−
Our approach for using the landmarks is to manually mark the distinct anatomical features by point landmarks after the automatic registration is done. Matching of landmarks is performed by two different scenarios. In one the optimization procedure is applied to the cost function containing an intensity-based data term and a constraint of explicit landmark matching. In another, an initial deformation field obtained from automatic registration is corrected by the field built on RBFs around the landmarks. The main challenge to this approach is <foo>.
+
Our approach for using the landmarks is to manually mark the distinct anatomical features by point landmarks after the automatic registration is done. Matching of landmarks is performed by two different methods. In one the optimization procedure is applied to the cost function containing an intensity-based data term and a landmark term, sum of squares of distances between fixed and moving landmarks. In another, an initial deformation field obtained from automatic registration is corrected by the field built on RBFs around the landmarks.  
  
 
Our plan for the project week is to first try out <bar>,...
 
Our plan for the project week is to first try out <bar>,...
Line 38: Line 38:
  
 
<h3>Progress</h3>
 
<h3>Progress</h3>
−
Software for the fiber tracking and statistical analysis along the tracts has been implemented. The statistical methods for diffusion tensors are implemented as ITK code as part of the [[NA-MIC/Projects/Diffusion_Image_Analysis/DTI_Software_and_Algorithm_Infrastructure|DTI Software Infrastructure]] project. The methods have been validated on a repeated scan of a healthy individual. This work has been published as a conference paper (MICCAI 2005) and a journal version (MEDIA 2006). Our recent IPMI 2007 paper includes a nonparametric regression method for analyzing data along a fiber tract.
+
The code for the two methods of landmark matching has been implemented to our in house open source Plastimatch software. The methods have been validated on synthetic images and on lung CT scans obtained from NBIA data base. Statistical analysis for method validation has been performed on 4D CT scans obtained from http://www.dir-lab.com
−
 
 
−
 
 
 
</div>
 
</div>
 
</div>
 
</div>
Line 55: Line 53:
 
##Extension -- commandline
 
##Extension -- commandline
 
##Extension -- loadable
 
##Extension -- loadable
−
#Other (Please specify)
+
#Other (YES) -- stand alone utility
  
 
==References==
 
==References==

Latest revision as of 19:11, 6 June 2010

Home < 2010 Summer Project Week Fiducial Deformable Registration

  • Top: Reference and test synthetic images. Bottom: Results for registration without landmark information and using one landmark.

  • Registration error vs number of landmarks.


Key Investigators

  • MGH: Nadya Shusharina, Greg Sharp

Objective

We are developing methods for deformable image registration with landmark information. The goal is to improve the registration if misaligned areas are found.




Approach, Plan

Our approach for using the landmarks is to manually mark the distinct anatomical features by point landmarks after the automatic registration is done. Matching of landmarks is performed by two different methods. In one the optimization procedure is applied to the cost function containing an intensity-based data term and a landmark term, sum of squares of distances between fixed and moving landmarks. In another, an initial deformation field obtained from automatic registration is corrected by the field built on RBFs around the landmarks.

Our plan for the project week is to first try out <bar>,...

Progress

The code for the two methods of landmark matching has been implemented to our in house open source Plastimatch software. The methods have been validated on synthetic images and on lung CT scans obtained from NBIA data base. Statistical analysis for method validation has been performed on 4D CT scans obtained from http://www.dir-lab.com

Delivery Mechanism

This work will be delivered to the NA-MIC Kit as a (please select the appropriate options by noting YES against them below)

  1. ITK Module
  2. Slicer Module
    1. Built-in
    2. Extension -- commandline
    3. Extension -- loadable
  3. Other (YES) -- stand alone utility

References

  • G. C. Sharp, R. Li, J. Wolfgang, G. TY Chen, M. Peroni, M. F. Spadea, S. Mori, J. Zhang, J. Shackleford, N. Kandasamy, Plastimatch - an open source software suite for radiotherapy image processing, In the proceedings of the 16th International Conference on the Use of Computers in Radiotherapy (ICCR 2010), Amsterdam, The Netherlands.
  • Plastimatch. http://plastimatch.org
Retrieved from "https://www.na-mic.org/w/index.php?title=2010_Summer_Project_Week_Fiducial_Deformable_Registration&oldid=53391"