MICCAI 2012 Grand Challenge and Workshop on Multi-Atlas Labeling
Characterization of anatomical structure through segmentation has become essential for morphological assessment and localizing quantitative measures. Segmentation through registration and atlas label transfer has proven to be a flexible and fruitful approach as efficient, non-rigid image registration methods have become prevalent. Label transfer segmentation using multiple atlases has helped to bring statistical fusion, shape modeling, and meta-analysis techniques to the forefront of segmentation research.
Numerous creative approaches have proposed to use atlas information to apply labels to brain anatomy. However, it is difficult to evaluate the relative advantages and limitations of these methods as they have been applied on very different datasets. This workshop will provide a snapshot of the current progress in the field through extended discussions and provide researchers an opportunity to characterize their methods on standardized data in a grand challenge. The full training/testing datasets will be released for community use after the competition of this challenge.
Multi-atlas techniques are increasingly becoming integral to numerous medical image computing approaches and for diverse anatomical regions. The proposed workshop will present a forum to discuss recent advances from multiple perspectives. The discussion will focus on new methods/applications of multi-atlas labeling and comparative evaluations of methods on a newly public, manually labeled dataset. First, we will host a discussion on methods, theory, and applications in which traditional MICCAI format papers will be solicited. Second, we will host a challenge on whole-brain labeling which will provide an opportunity to characterize multiple labels, inter-label relationships, and heterogeneous structures (sponsored by NeuroMorphometrics). Submissions will be encouraged using a limited 3-4 page format.
Topics of Interest
Topics of interests include but are not limited to:
- Multi-atlas registration
- Statistical methods for label fusion
- Theory and applications with discrete, continuous, or non-traditional label types
- Manifolds theory and applications for voxel, volume, surface, or non-traditional multi-atlas representations
- Atlas design, selection, and exclusion
- Multi-atlas informed and augmented approaches, including shape modeling
- Applications of multi-atlas methods for segmentation and labeling
- Visualization and hypothesis exploration approaches using multi-atlases