Difference between revisions of "2013 Summer Project Week:SegmentationAidedRegistration"

Revision as of 21:17, 18 June 2013

Home < 2013 Summer Project Week:SegmentationAidedRegistration

Projects List
Two images (3D) to be registered.
Segmentation aided registration using this module.
Segmentation aided registration using this module, with red contour of left atrium.
MI registration.

Yi Gao, LiangJia Zhu, Josh Cates, Rob MacLeod, Sylvain Bouix, Ron Kikinis, Allen Tannenbaum

When registering two images, sometime we want some specific regions in the two images to be particularly accurately matched.

We use the segmentation of the target region to aided the registration.

Extension in nightly build. Can be installed from Extension manager.

This work will be delivered to the NA-MIC Kit as a commandline extension.

Utah DBP
Y. Gao and S. Bouix, “Synthesis of realistic subcortical anatomy with known surface deformations,” in MICCAI Workshop on Mesh Processing in Medical Image Analysis, 2012, pp. 80–88.
T. Riklin-Raviv, Y. Gao, J. Levitt, and S. Bouix, “Statistical shape analysis for population studies via level-set based shape morphing,” in ECCV Workshop on Non-Rigid Shape Analysis and Deformable Image Alignment, 2012, pp. 42–51.

@@ Line 16: / Line 16: @@
 <h3>Objective</h3>
-Among the AFib patients underwent RF ablation, the relative high AFib recurrence rate is a concern. The correlation between the cure/recurrence ratio with the distribution of the fibrosis would provide insight on the disease assessment and treatment planning.
+When registering two images, sometime we want some specific regions in the two images to be particularly accurately matched.
 </div>
@@ Line 24: / Line 23: @@
 <h3>Approach, Plan</h3>
-The fibrosis distributions on the left atrium wall is imaged using the dynamic contrast enhanced MRI. Distributed on different anatomical structures, they are considered as "mass" defined on different domains. Under the framework of the optimal mass transport (OMT), the masses are transported to a common domain where the statistical analysis can then be applied. The significant different regions are then characterized by the low-p-value regions.
+We use the segmentation of the target region to aided the registration.
 </div>
@@ Line 31: / Line 29: @@
 <h3>Progress</h3>
-C++ code using ITK and VTK is finished.
+Extension in nightly build. Can be installed from Extension manager.