Difference between revisions of "NA-MIC NCBC Collaboration:Development and Dissemination of Robust Brain MRI Measurement Tools"
From NAMIC Wiki
m |
|||
| (4 intermediate revisions by 2 users not shown) | |||
| Line 1: | Line 1: | ||
| − | Back to [[NA-MIC_External_Collaborations]] | + | Back to [[NA-MIC_External_Collaborations|NA-MIC External Collaborations]] |
==Abstract== | ==Abstract== | ||
| Line 6: | Line 6: | ||
==Grant#== | ==Grant#== | ||
| − | |||
R01EB006733 | R01EB006733 | ||
| − | + | ==Funding Duration== | |
| + | 09/17/2008-08/31/2011 | ||
==Key Personnel== | ==Key Personnel== | ||
| − | + | * UNC: Dinggang Shen, PI | |
| − | Dinggang Shen, Xiaodong Tao, | + | * NA-MIC: Jim Miller (GE Research, Investigator), Xiaodong Tao (GE Research, Investigator) |
| + | * Christos Davatzikos (UPenn, Consultant) | ||
==Projects== | ==Projects== | ||
| − | |||
* Perform various novel algorithm improvements and developments, aiming at significantly enhancing the robustness of HAMMER registration and WML segmentation algorithms to a wide variety of levels of image quality and type. In particular, the HAMMER registration algorithm will be extended to leverage training datasets to improve the automated anatomical feature identification and matching during the image registration procedure. This will allow the algorithm to be adapted to the specific characteristics of different scanners, imaging protocols, and studies via a procedure that is analogous to the training of human experts. WML segmentation algorithm will be extended to learn the spatial distribution of WML from the training datasets, assisting WML segmentation. Also, the robustness of the multimodality image registration in the WML segmentation algorithm will be improved by a quantitative-qualitative measure of mutual information (MI) that integrates directly the image spatial information into the MI calculation. | * Perform various novel algorithm improvements and developments, aiming at significantly enhancing the robustness of HAMMER registration and WML segmentation algorithms to a wide variety of levels of image quality and type. In particular, the HAMMER registration algorithm will be extended to leverage training datasets to improve the automated anatomical feature identification and matching during the image registration procedure. This will allow the algorithm to be adapted to the specific characteristics of different scanners, imaging protocols, and studies via a procedure that is analogous to the training of human experts. WML segmentation algorithm will be extended to learn the spatial distribution of WML from the training datasets, assisting WML segmentation. Also, the robustness of the multimodality image registration in the WML segmentation algorithm will be improved by a quantitative-qualitative measure of mutual information (MI) that integrates directly the image spatial information into the MI calculation. | ||
| Line 22: | Line 22: | ||
==Publications== | ==Publications== | ||
| − | * Dinggang Shen, Christos Davatzikos, [http://www.med.unc.edu/~dgshen/papers/Hammer_VersionInTMI.pdf HAMMER: Hierarchical Attribute Matching Mechanism for Elastic Registration] | + | * Dinggang Shen, Christos Davatzikos, [http://www.med.unc.edu/~dgshen/papers/Hammer_VersionInTMI.pdf HAMMER: Hierarchical Attribute Matching Mechanism for Elastic Registration.]IEEE Trans Med Imaging. 2002 Nov;21(11):1421-39. PMID: 12575879. |
| − | + | * Zhiqiang Lao, Dinggang Shen, Dengfeng Liu, Abbas F Jawad, Elias R Melhem, Lenore J Launer, Nick R Bryan, Christos Davatzikos, [http://www.med.unc.edu/~dgshen/papers/WMlesionSegmentation.pdf Computer-Assisted Segmentation of White Matter Lesions in 3D MR images, Using Pattern Recognition.] Acad Radiol. 2008 Mar;15(3):300-13. PMID: 18280928. | |
| − | * Zhiqiang Lao, Dinggang Shen, Dengfeng Liu, Abbas F Jawad, Elias R Melhem, Lenore J Launer, Nick R Bryan, Christos Davatzikos, [http://www.med.unc.edu/~dgshen/papers/WMlesionSegmentation.pdf Computer-Assisted Segmentation of White Matter Lesions in 3D MR images, Using Pattern Recognition] | ||
==External Resources== | ==External Resources== | ||
[http://www.med.unc.edu/~dgshen/HAMMER.htm HAMMER] | [http://www.med.unc.edu/~dgshen/HAMMER.htm HAMMER] | ||
Latest revision as of 19:06, 16 December 2009
Home < NA-MIC NCBC Collaboration:Development and Dissemination of Robust Brain MRI Measurement ToolsBack to NA-MIC External Collaborations
Contents
Abstract
This project aims at developing and widely distributing a software package for robust measurement of brain structure in MR images, via collaboration with the National Alliance for Medical Image Computing (NA-MIC) that will integrate this software into the 3D Slicer (currently being developing in NA-MIC). This particular software package will include a brain image registration and warping algorithm, called HAMMER, and an algorithm for computer-based segmentation of white matter lesions (WMLs), which can arise from a variety of pathologies including vascular pathology and multiple sclerosis.
Grant#
R01EB006733
Funding Duration
09/17/2008-08/31/2011
Key Personnel
- UNC: Dinggang Shen, PI
- NA-MIC: Jim Miller (GE Research, Investigator), Xiaodong Tao (GE Research, Investigator)
- Christos Davatzikos (UPenn, Consultant)
Projects
- Perform various novel algorithm improvements and developments, aiming at significantly enhancing the robustness of HAMMER registration and WML segmentation algorithms to a wide variety of levels of image quality and type. In particular, the HAMMER registration algorithm will be extended to leverage training datasets to improve the automated anatomical feature identification and matching during the image registration procedure. This will allow the algorithm to be adapted to the specific characteristics of different scanners, imaging protocols, and studies via a procedure that is analogous to the training of human experts. WML segmentation algorithm will be extended to learn the spatial distribution of WML from the training datasets, assisting WML segmentation. Also, the robustness of the multimodality image registration in the WML segmentation algorithm will be improved by a quantitative-qualitative measure of mutual information (MI) that integrates directly the image spatial information into the MI calculation.
- Develop an easy-to-use, robust software module for the HAMMER registration algorithm and incorporate this module into the 3D Slicer. The robustness and performance of this module will be tested using the NA-MIC Software Engineering Process which includes the Dart2 Software Testing System, as well as by manual testing by clinical experts, and using various clinical research data acquired from different imaging centers. The software module will expose access to the parameters of the HAMMER algorithm at a variety of levels to support the variety of needs of different users.
- Develop an easy-to-use, robust software module for WML segmentation algorithm, and incorporate this module into the 3D Slicer. Users will have the choice of a pre-trained model for WML segmentation or directly training the algorithm using their own data. The robustness and performance of this module will be tested using the NA-MIC Software Engineering Process which includes the Dart2 Software Testing System, as well as manual testing by clinical experts.
Publications
- Dinggang Shen, Christos Davatzikos, HAMMER: Hierarchical Attribute Matching Mechanism for Elastic Registration.IEEE Trans Med Imaging. 2002 Nov;21(11):1421-39. PMID: 12575879.
- Zhiqiang Lao, Dinggang Shen, Dengfeng Liu, Abbas F Jawad, Elias R Melhem, Lenore J Launer, Nick R Bryan, Christos Davatzikos, Computer-Assisted Segmentation of White Matter Lesions in 3D MR images, Using Pattern Recognition. Acad Radiol. 2008 Mar;15(3):300-13. PMID: 18280928.
