National Alliance for Medical Image Computing

The Publication Database hosted by SPL

All Publications | Upload | Advanced Search | Gallery View | Download Statistics | Help | Import | Log in

Optimal Parameter Map Estimation for Shape Representation: A Generative Approach

Institution:
Scientific Computing and Imaging Institute, University of Utah, Salt Lake City, UT, USA.
Publication Date:
Apr-2016
Journal:
Proc IEEE Int Symp Biomed Imaging
Volume Number:
2016
Pages:
660-3
Citation:
Proc IEEE Int Symp Biomed Imaging. 2016 Apr;2016:660-3.
PubMed ID:
28090247
PMCID:
PMC5228593
Keywords:
consensus generation, generative models, parameter map, probabilistic labeling, shape representation
Appears in Collections:
NA-MIC
Sponsors:
P41 GM103545/GM/NIGMS NIH HHS/United States
U54 EB005149/EB/NIBIB NIH HHS/United States
Generated Citation:
Elhabian S.Y., Agrawal P., Whitaker R.T. Optimal Parameter Map Estimation for Shape Representation: A Generative Approach. Proc IEEE Int Symp Biomed Imaging. 2016 Apr;2016:660-3. PMID: 28090247. PMCID: PMC5228593.
Downloaded: 38 times. [view map]
Paper: Download, View online
Export citation:
Google Scholar: link

Probabilistic label maps are a useful tool for important medical image analysis tasks such as segmentation, shape analysis, and atlas building. Existing methods typically rely on blurred signed distance maps or smoothed label maps to model uncertainties and shape variabilities, which do not conform to any generative model or estimation process, and are therefore suboptimal. In this paper, we propose to learn probabilistic label maps using a generative model on given set of binary label maps. The proposed approach generalizes well on unseen data while simultaneously capturing the variability in the training samples. Efficiency of the proposed approach is demonstrated for consensus generation and shape-based clustering using synthetic datasets as well as left atrial segmentations from late-gadolinium enhancement MRI.

Additional Material
1 File (230.275kB)
Elhabian-ISBI2016-fig5.jpg (230.275kB)