Difference between revisions of "Algorithm:GATech"

From NAMIC Wiki
Jump to: navigation, search
Line 5: Line 5:
 
At Georgia Tech, we are broadly interested in a range of mathematical image analysis algorithms for segmentation, registration, diffusion-weighted MRI analysis, and statistical analysis.  For many applications, we cast the problem in an energy minimization framework wherein we define a partial differential equation whose numeric solution corresponds to the desired algorithmic outcome.  The following are many examples of PDE techniques applied to medical image analysis.
 
At Georgia Tech, we are broadly interested in a range of mathematical image analysis algorithms for segmentation, registration, diffusion-weighted MRI analysis, and statistical analysis.  For many applications, we cast the problem in an energy minimization framework wherein we define a partial differential equation whose numeric solution corresponds to the desired algorithmic outcome.  The following are many examples of PDE techniques applied to medical image analysis.
  
== Segmentation ==
+
= Georgia Tech Projects =
  
 
{|
 
{|
Line 11: Line 11:
 
| style="width:90%" |
 
| style="width:90%" |
  
=== [[Algorithm:GATech:Multiscale_Shape_Segmentation|Multiscale Shape Segmentation Techniques]] ===
+
== [[Algorithm:GATech:Multiscale_Shape_Segmentation|Multiscale Shape Segmentation Techniques]] ==
  
 
To represent multiscale variations in a shape population in order to drive the segmentation of deep brain structures, such as the caudate nucleus or the hippocampus.
 
To represent multiscale variations in a shape population in order to drive the segmentation of deep brain structures, such as the caudate nucleus or the hippocampus.
Line 19: Line 19:
 
|-
 
|-
  
| | [[Image:Striatum1.png|thumb|left|200px|]]
+
| | [[Image:ZoomedResultWithModel.png|thumb|left|200px]]
 
| |
 
| |
  
=== [[Algorithm:GATech:Rule_Based_Segmentation|Rule-Based Segmentation Techniques]] ===
+
== [[Algorithm:GATech:DWMRI_Geodesic_Active_Contours|Geodesic Active Contours for Fiber Tractography and Fiber Bundle Segmentation]] ==
  
In this work, we provide software to semi-automate the implementation of segmentation procedures based on expert neuroanatomist rules.
+
In this work, we provide an energy minimization framework which allows one to find fiber tracts and volumetric fiber bundles in brain diffusion-weighted MRI (DW-MRI).
  
<font color="red">'''New: '''</font> Al-Hakim, et al. Parcellation of the Striatum. SPIE MI 2007.
+
<font color="red">'''New: '''</font> J. Melonakos, E. Pichon, S. Angenet, and A. Tannenbaum. Finsler Active Contours. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007 (in press).
  
 
|-
 
|-
  
| | [[Image:Circle seg.PNG|thumb|left|200px|]]
+
| | [[Image:Results brain sag.JPG|thumb|left|200px|]]
 
| |
 
| |
  
=== [[Algorithm:GATech:KPCA_Segmentation|Kernel PCA for Segmentation]] ===
+
== [[Algorithm:GATech:Optimal_Mass_Transport_Registration|Optimal Mass Transport Registration]] ==
  
Segmentation performances using active contours can be drastically improved if the possible shapes of the object of interest are learnt. The goal of this work is to use Kernel PCA to learn shape priors. Kernel PCA allows for learning non linear dependencies in data sets, leading to more robust shape priors.
+
The goal of this project is to implement a computationaly efficient Elastic/Non-rigid Registration algorithm based on the Monge-Kantorovich theory of optimal mass transport for 3D Medical Imagery. Our technique is based on Multigrid and Multiresolution techniques. This method is particularly useful because it is parameter free and utilizes all of the grayscale data in the image pairs in a symmetric fashion and no landmarks need to be specified for correspondence.
  
<font color="red">'''New: '''</font> S. Dambreville, Y. Rathi, and A. Tannenbaum. A Framework for Image Segmentation using Image Shape Models and Kernel PCA Shape Priors. PAMI. Submitted to PAMI.
+
<font color="red">'''New: '''</font> Tauseef ur Rehman, A. Tannenbaum. Multigrid Optimal Mass Transport for Image Registration and Morphing. SPIE Conference on Computational Imaging V, Jan 2007.
  
 
|-
 
|-
  
| | [[Image:Fig1yan.PNG|thumb|left|200px|]]
+
| | [[Image:Striatum1.png|thumb|left|200px|]]
 
| |
 
| |
  
=== [[Algorithm:GATech:Blood_Vessel_Segmentation|Blood Vessel Segmentation]] ===
+
== [[Algorithm:GATech:Rule_Based_Segmentation|Rule-Based Segmentation Techniques]] ==
  
The goal of this work is to develop blood vessel segmentation techniques for 3D MRI and CT data. The methods have been applied to coronary arteries and portal veins, with promising results.
+
In this work, we provide software to semi-automate the implementation of segmentation procedures based on expert neuroanatomist rules.
  
<font color="red">'''New: '''</font>Y. Yang, S. George, D. Martin, A. Tannenbaum, and D. Giddens. 3D Modeling of Patient-Specific Geometries of Portal Veins Using MR Images. In Proceedings IEEE EMBS, 2006
+
<font color="red">'''New: '''</font> Al-Hakim, et al. Parcellation of the Striatum. SPIE MI 2007.
  
 
|-
 
|-
  
| | [[Image:Fig67.png|thumb|left|200px|]]
+
| | [[Image:Brain-flat.PNG|thumb|left|200px]]
 
| |
 
| |
  
=== [[Algorithm:GATech:Knowledge_Based_Bayesian_Segmentation|Knowledge-Based Bayesian Segmentation]] ===
+
== [[Algorithm:GATech:Conformal_Flattening_Registration|Conformal Flattening]] ==
  
This ITK filter is a segmentation algorithm that utilizes Bayes's Rule along with an affine-invariant anisotropic smoothing filter.
+
The goal of this project is for better visualizing and computation of neural activity from fMRI brain imagery. Also, with this technique, shapes can be mapped to shperes for shape analysis, registration or other purposes. Our technique is based on conformal mappings which map genus-zero surface: in fmri case cortical or other surfaces, onto a sphere in an angle preserving manner.
  
<font color="red">'''New: '''</font> J. Melonakos, Y. Gao, and A. Tannenbaum. Tissue Tracking: Applications for Brain MRI Classification. SPIE Medical Imaging, 2007.
+
<font color="red">'''New: '''</font> Y. Gao, J. Melonakos, and A. Tannenbaum. Conformal Flattening ITK Filter. ISC/NA-MIC Workshop on Open Science at MICCAI 2006.
  
 
|-
 
|-
  
| | [[Image:Stochastic-snake.png|thumb|left|200px|]]
+
| | [[Image:Basis membership.png|thumb|left|200px]]
 
| |
 
| |
  
=== [[Algorithm:GATech:Stochastic_Methods_Segmentation|Stochastic Methods for Segmentation]] ===
+
== [[Algorithm:GATech:Multiscale_Shape_Analysis|Multiscale Shape Analysis]] ==
  
New stochastic methods for implementing curvature driven flows for various medical tasks such as segmentation.
+
We present a novel method of statistical surface-based morphometry based on the use of non-parametric permutation tests and a spherical wavelet (SWC) shape representation.
  
<font color="red">'''New: '''</font> Currently under investigation.
+
<font color="red">'''New: '''</font> D. Nain, M. Styner, M. Niethammer, J. J. Levitt, M E Shenton, G Gerig, A. Bobick, A. Tannenbaum. Statistical Shape Analysis of Brain Structures using Spherical Wavelets. Accepted in The Fourth IEEE International Symposium on Biomedical Imaging (ISBI ’07) that will be held April 12-15, 2007 in Metro Washington DC, USA.
|}
 
  
== Registration ==
+
|-
  
 +
| | [[Image:Circle seg.PNG|thumb|left|200px|]]
 +
| |
  
{|
+
== [[Algorithm:GATech:KPCA_Segmentation|Kernel PCA for Segmentation]] ==
| style="width:10%" | [[Image:Brain-flat.PNG|thumb|left|200px]]
 
| style="width:90%" |
 
  
=== [[Algorithm:GATech:Conformal_Flattening_Registration|Conformal Flattening]] ===
+
Segmentation performances using active contours can be drastically improved if the possible shapes of the object of interest are learnt. The goal of this work is to use Kernel PCA to learn shape priors. Kernel PCA allows for learning non linear dependencies in data sets, leading to more robust shape priors.
  
The goal of this project is for better visualizing and computation of neural activity from fMRI brain imagery. Also, with this technique, shapes can be mapped to shperes for shape analysis, registration or other purposes. Our technique is based on conformal mappings which map genus-zero surface: in fmri case cortical or other surfaces, onto a sphere in an angle preserving manner.
+
<font color="red">'''New: '''</font> S. Dambreville, Y. Rathi, and A. Tannenbaum. A Framework for Image Segmentation using Image Shape Models and Kernel PCA Shape Priors. PAMI. Submitted to PAMI.
 
 
<font color="red">'''New: '''</font> Y. Gao, J. Melonakos, and A. Tannenbaum. Conformal Flattening ITK Filter. ISC/NA-MIC Workshop on Open Science at MICCAI 2006.
 
  
 
|-
 
|-
  
| | [[Image:Results brain sag.JPG|thumb|left|200px|]]
+
| | [[Image:Fig1yan.PNG|thumb|left|200px|]]
 
| |
 
| |
  
=== [[Algorithm:GATech:Optimal_Mass_Transport_Registration|Optimal Mass Transport Registration]] ===
+
== [[Algorithm:GATech:Blood_Vessel_Segmentation|Blood Vessel Segmentation]] ==
  
The goal of this project is to implement a computationaly efficient Elastic/Non-rigid Registration algorithm based on the Monge-Kantorovich theory of optimal mass transport for 3D Medical Imagery. Our technique is based on Multigrid and Multiresolution techniques. This method is particularly useful because it is parameter free and utilizes all of the grayscale data in the image pairs in a symmetric fashion and no landmarks need to be specified for correspondence.
+
The goal of this work is to develop blood vessel segmentation techniques for 3D MRI and CT data. The methods have been applied to coronary arteries and portal veins, with promising results.
  
<font color="red">'''New: '''</font> Tauseef ur Rehman, A. Tannenbaum. Multigrid Optimal Mass Transport for Image Registration and Morphing. SPIE Conference on Computational Imaging V, Jan 2007.
+
<font color="red">'''New: '''</font>Y. Yang, S. George, D. Martin, A. Tannenbaum, and D. Giddens. 3D Modeling of Patient-Specific Geometries of Portal Veins Using MR Images. In Proceedings IEEE EMBS, 2006
|}
 
  
== DW-MRI Processing ==
+
|-
  
{|
+
| | [[Image:Fig67.png|thumb|left|200px|]]
| style="width:10%" | [[Image:ZoomedResultWithModel.png|thumb|left|200px]]
+
| |
| style="width:90%" |
 
  
=== [[Algorithm:GATech:DWMRI_Geodesic_Active_Contours|Geodesic Active Contours for Fiber Tractography and Fiber Bundle Segmentation]] ===
+
== [[Algorithm:GATech:Knowledge_Based_Bayesian_Segmentation|Knowledge-Based Bayesian Segmentation]] ==
  
In this work, we provide an energy minimization framework which allows one to find fiber tracts and volumetric fiber bundles in brain diffusion-weighted MRI (DW-MRI).
+
This ITK filter is a segmentation algorithm that utilizes Bayes's Rule along with an affine-invariant anisotropic smoothing filter.
  
<font color="red">'''New: '''</font> J. Melonakos, E. Pichon, S. Angenet, and A. Tannenbaum. Finsler Active Contours. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007 (in press).
+
<font color="red">'''New: '''</font> J. Melonakos, Y. Gao, and A. Tannenbaum. Tissue Tracking: Applications for Brain MRI Classification. SPIE Medical Imaging, 2007.
|}
 
  
== Shape Analysis ==
+
|-
  
{|
+
| | [[Image:Stochastic-snake.png|thumb|left|200px|]]
| style="width:10%" | [[Image:Basis membership.png|thumb|left|200px]]
+
| |
| style="width:90%" |
 
  
=== [[Algorithm:GATech:Multiscale_Shape_Analysis|Multiscale Shape Analysis]] ===
+
== [[Algorithm:GATech:Stochastic_Methods_Segmentation|Stochastic Methods for Segmentation]] ==
  
We present a novel method of statistical surface-based morphometry based on the use of non-parametric permutation tests and a spherical wavelet (SWC) shape representation.
+
New stochastic methods for implementing curvature driven flows for various medical tasks such as segmentation.
  
<font color="red">'''New: '''</font> D. Nain, M. Styner, M. Niethammer, J. J. Levitt, M E Shenton, G Gerig, A. Bobick, A. Tannenbaum. Statistical Shape Analysis of Brain Structures using Spherical Wavelets. Accepted in The Fourth IEEE International Symposium on Biomedical Imaging (ISBI ’07) that will be held April 12-15, 2007 in Metro Washington DC, USA.
+
<font color="red">'''New: '''</font> Currently under investigation.
  
 
|-
 
|-
Line 128: Line 121:
 
| |
 
| |
  
=== [[Algorithm:GATech:KPCA_LLE_KLLE_Shape_Analysis|KPCA, LLE, KLLE Shape Analysis]] ===
+
== [[Algorithm:GATech:KPCA_LLE_KLLE_Shape_Analysis|KPCA, LLE, KLLE Shape Analysis]] ==
  
 
The goal of this work is to study and compare shape learning techniques. The techniques considered are Linear Principal Components Analysis (PCA), Kernel PCA, Locally Linear Embedding (LLE) and Kernel LLE.
 
The goal of this work is to study and compare shape learning techniques. The techniques considered are Linear Principal Components Analysis (PCA), Kernel PCA, Locally Linear Embedding (LLE) and Kernel LLE.
  
 
<font color="red">'''New: '''</font>  Y. Rathi, S. Dambreville, and A. Tannenbaum. "Comparative Analysis of Kernel Methods for Statistical Shape Learning", In CVAMIA held in conjunction with ECCV, 2006.
 
<font color="red">'''New: '''</font>  Y. Rathi, S. Dambreville, and A. Tannenbaum. "Comparative Analysis of Kernel Methods for Statistical Shape Learning", In CVAMIA held in conjunction with ECCV, 2006.
|}
 
  
== Segmentation ==
+
|-
  
{|
+
| | [[Image:Gatech SlicerModel2.jpg|thumb|left|200px]]
| style="width:10%" | [[Image:Gatech SlicerModel2.jpg|thumb|left|200px]]
+
| |
| style="width:90%" |
 
  
=== [[Algorithm:GATech:Statistical_Segmentation_Slicer_2|Statistical/PDE Methods using Fast Marching for Segmentation]] ===
+
== [[Algorithm:GATech:Statistical_Segmentation_Slicer_2|Statistical/PDE Methods using Fast Marching for Segmentation]] ==
  
 
This Fast Marching based flow was added to Slicer 2.
 
This Fast Marching based flow was added to Slicer 2.
Line 150: Line 141:
 
| |
 
| |
  
=== [[Algorithm:GATech:Image_Smooth_Slicer_2|Image Smooth Slicer 2 Module]] ===
+
== [[Algorithm:GATech:Image_Smooth_Slicer_2|Image Smooth Slicer 2 Module]] ==
  
 
2D and 3D smoothing of images.
 
2D and 3D smoothing of images.
Line 159: Line 150:
 
| |
 
| |
  
=== [[Algorithm:Affine_Segment_Slicer_2|Affine Segment Slicer 2 Module]] ===
+
== [[Algorithm:Affine_Segment_Slicer_2|Affine Segment Slicer 2 Module]] ==
  
 
This module can be used to perform edge based segmentation.
 
This module can be used to perform edge based segmentation.
 
|}
 
|}

Revision as of 19:42, 16 August 2007

Home < Algorithm:GATech
Back to NA-MIC Algorithms

Overview of Georgia Tech Algorithms

At Georgia Tech, we are broadly interested in a range of mathematical image analysis algorithms for segmentation, registration, diffusion-weighted MRI analysis, and statistical analysis. For many applications, we cast the problem in an energy minimization framework wherein we define a partial differential equation whose numeric solution corresponds to the desired algorithmic outcome. The following are many examples of PDE techniques applied to medical image analysis.

Georgia Tech Projects

Gatech caudateBands.PNG

Multiscale Shape Segmentation Techniques

To represent multiscale variations in a shape population in order to drive the segmentation of deep brain structures, such as the caudate nucleus or the hippocampus.

New: Delphine Nain won the best student paper at MICCAI 2006 in the category "Segmentation and Registration" for her paper entitled "Shape-driven surface segmentation using spherical wavelets" by D. Nain, S. Haker, A. Bobick, A. Tannenbaum.

ZoomedResultWithModel.png

Geodesic Active Contours for Fiber Tractography and Fiber Bundle Segmentation

In this work, we provide an energy minimization framework which allows one to find fiber tracts and volumetric fiber bundles in brain diffusion-weighted MRI (DW-MRI).

New: J. Melonakos, E. Pichon, S. Angenet, and A. Tannenbaum. Finsler Active Contours. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007 (in press).

Results brain sag.JPG

Optimal Mass Transport Registration

The goal of this project is to implement a computationaly efficient Elastic/Non-rigid Registration algorithm based on the Monge-Kantorovich theory of optimal mass transport for 3D Medical Imagery. Our technique is based on Multigrid and Multiresolution techniques. This method is particularly useful because it is parameter free and utilizes all of the grayscale data in the image pairs in a symmetric fashion and no landmarks need to be specified for correspondence.

New: Tauseef ur Rehman, A. Tannenbaum. Multigrid Optimal Mass Transport for Image Registration and Morphing. SPIE Conference on Computational Imaging V, Jan 2007.

Striatum1.png

Rule-Based Segmentation Techniques

In this work, we provide software to semi-automate the implementation of segmentation procedures based on expert neuroanatomist rules.

New: Al-Hakim, et al. Parcellation of the Striatum. SPIE MI 2007.

Brain-flat.PNG

Conformal Flattening

The goal of this project is for better visualizing and computation of neural activity from fMRI brain imagery. Also, with this technique, shapes can be mapped to shperes for shape analysis, registration or other purposes. Our technique is based on conformal mappings which map genus-zero surface: in fmri case cortical or other surfaces, onto a sphere in an angle preserving manner.

New: Y. Gao, J. Melonakos, and A. Tannenbaum. Conformal Flattening ITK Filter. ISC/NA-MIC Workshop on Open Science at MICCAI 2006.

Basis membership.png

Multiscale Shape Analysis

We present a novel method of statistical surface-based morphometry based on the use of non-parametric permutation tests and a spherical wavelet (SWC) shape representation.

New: D. Nain, M. Styner, M. Niethammer, J. J. Levitt, M E Shenton, G Gerig, A. Bobick, A. Tannenbaum. Statistical Shape Analysis of Brain Structures using Spherical Wavelets. Accepted in The Fourth IEEE International Symposium on Biomedical Imaging (ISBI ’07) that will be held April 12-15, 2007 in Metro Washington DC, USA.

Circle seg.PNG

Kernel PCA for Segmentation

Segmentation performances using active contours can be drastically improved if the possible shapes of the object of interest are learnt. The goal of this work is to use Kernel PCA to learn shape priors. Kernel PCA allows for learning non linear dependencies in data sets, leading to more robust shape priors.

New: S. Dambreville, Y. Rathi, and A. Tannenbaum. A Framework for Image Segmentation using Image Shape Models and Kernel PCA Shape Priors. PAMI. Submitted to PAMI.

Fig1yan.PNG

Blood Vessel Segmentation

The goal of this work is to develop blood vessel segmentation techniques for 3D MRI and CT data. The methods have been applied to coronary arteries and portal veins, with promising results.

New: Y. Yang, S. George, D. Martin, A. Tannenbaum, and D. Giddens. 3D Modeling of Patient-Specific Geometries of Portal Veins Using MR Images. In Proceedings IEEE EMBS, 2006

Fig67.png

Knowledge-Based Bayesian Segmentation

This ITK filter is a segmentation algorithm that utilizes Bayes's Rule along with an affine-invariant anisotropic smoothing filter.

New: J. Melonakos, Y. Gao, and A. Tannenbaum. Tissue Tracking: Applications for Brain MRI Classification. SPIE Medical Imaging, 2007.

Stochastic-snake.png

Stochastic Methods for Segmentation

New stochastic methods for implementing curvature driven flows for various medical tasks such as segmentation.

New: Currently under investigation.

Table1.png

KPCA, LLE, KLLE Shape Analysis

The goal of this work is to study and compare shape learning techniques. The techniques considered are Linear Principal Components Analysis (PCA), Kernel PCA, Locally Linear Embedding (LLE) and Kernel LLE.

New: Y. Rathi, S. Dambreville, and A. Tannenbaum. "Comparative Analysis of Kernel Methods for Statistical Shape Learning", In CVAMIA held in conjunction with ECCV, 2006.

Gatech SlicerModel2.jpg

Statistical/PDE Methods using Fast Marching for Segmentation

This Fast Marching based flow was added to Slicer 2.

Image Smooth Slicer 2 Module

2D and 3D smoothing of images.

Affine Segment Slicer 2 Module

This module can be used to perform edge based segmentation.