Difference between revisions of "Projects:RegistrationRegularization"

From NAMIC Wiki
Jump to: navigation, search
Line 1: Line 1:
Collaborators: B.T. Thomas Yeo (MIT), Mert Sabuncu (MIT), Rahul Desikan (BU), Polina Golland (MIT), Bruce Fischl (MGH).
+
Back to [[NA-MIC_Collaborations|NA-MIC_Collaborations]], [[Algorithm:MIT|MIT Algorithms]]
 +
 
 +
We propose a unified framework for computing atlases from manually labeled data sets at various degrees of “sharpness” and the joint registration and segmentation of a new brain with these atlases. In non-rigid registration, the tradeoff between warp regularization and image fidelity is typically set empirically. In segmentation, this leads to a probabilistic atlas of arbitrary “sharpness”: weak regularization results in well-aligned training images, producing a “sharp” atlas; strong regularization yields a “blurry” atlas. We study the effects of this tradeoff in the context of cortical surface parcellation, but the framework applies to volume registration as well.  
  
 
= Description =
 
= Description =
We propose a unified framework for computing atlases from manually labeled data sets at various degrees of “sharpness” and the joint registration and segmentation of a new brain with these atlases. In non-rigid registration, the tradeoff between warp regularization and image fidelity is typically set empirically. In segmentation, this leads to a probabilistic atlas of arbitrary “sharpness”: weak regularization results in well-aligned training images, producing a “sharp” atlas; strong regularization yields a “blurry” atlas. We study the effects of this tradeoff in the context of cortical surface parcellation, but the framework applies to volume registration as well.
 
  
= Results =
 
 
We compare three special cases of our framework, namely:  
 
We compare three special cases of our framework, namely:  
  
Line 20: Line 20:
 
[[Image:AvgResults.jpg|thumb|center|400px|Plot of Dice as a function of the warp scale S. Note that S is on a log scale. <math>\alpha</math> corresponds to the sharpness of the atlas used.]]
 
[[Image:AvgResults.jpg|thumb|center|400px|Plot of Dice as a function of the warp scale S. Note that S is on a log scale. <math>\alpha</math> corresponds to the sharpness of the atlas used.]]
  
= References =
+
= Key Investigator =
 +
 
 +
MIT: B.T. Thomas Yeo, Mert Sabuncu, Polina Golland
 +
BU: Rahul Desikan
 +
MGH: Bruce Fischl
 +
 
 +
 
 +
= Publication =
 
[1] B. Fischl et al. Automatically Parcellating the Human cerebral Cortex. Cerebral
 
[1] B. Fischl et al. Automatically Parcellating the Human cerebral Cortex. Cerebral
 
Cortex, 14:11-22, 2004.
 
Cortex, 14:11-22, 2004.
Line 27: Line 34:
 
cerebral cortex on MRI scans into gyral based regions of interest. NeuroImage,
 
cerebral cortex on MRI scans into gyral based regions of interest. NeuroImage,
 
31:968-980, 2006.
 
31:968-980, 2006.
 +
 +
= Links =

Revision as of 01:14, 21 September 2007

Home < Projects:RegistrationRegularization
Back to NA-MIC_Collaborations, MIT Algorithms

We propose a unified framework for computing atlases from manually labeled data sets at various degrees of “sharpness” and the joint registration and segmentation of a new brain with these atlases. In non-rigid registration, the tradeoff between warp regularization and image fidelity is typically set empirically. In segmentation, this leads to a probabilistic atlas of arbitrary “sharpness”: weak regularization results in well-aligned training images, producing a “sharp” atlas; strong regularization yields a “blurry” atlas. We study the effects of this tradeoff in the context of cortical surface parcellation, but the framework applies to volume registration as well.

Description

We compare three special cases of our framework, namely:

(1) Progressive registration of a new brain to increasingly “sharp” atlases using increasingly flexible warps, by initializing each registration stage with the optimal warps from a “blurrier” atlas. We call this multiple atlases, multiple warp scales (MAMS).

(2) Progressive registration to a single atlas with increasingly flexible warps. We call this single atlas, multiple warp scales (SAMS).

(3) Registration to a single atlas with fixed constrained warps. We call this single atlas, single warp scale (SASS).

We use dice as the measure of segmentation quality. From the graph below, we note that the optimal parcellation algorithm in all three cases yield a statistically significant improvement over a state-of-the-art benchmark parcellation algorithm [1,2]. The optimal algorithms correspond to a unique balance between atlas “sharpness” and warp regularization.

Plot of Dice as a function of the warp scale S. Note that S is on a log scale. [math]\alpha[/math] corresponds to the sharpness of the atlas used.

Key Investigator

MIT: B.T. Thomas Yeo, Mert Sabuncu, Polina Golland BU: Rahul Desikan MGH: Bruce Fischl


Publication

[1] B. Fischl et al. Automatically Parcellating the Human cerebral Cortex. Cerebral Cortex, 14:11-22, 2004.

[2] R.S. Desikan et al. An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest. NeuroImage, 31:968-980, 2006.

Links