Intrinsic Shape Analysis
Using intrinsic geometry, in particular the Laplace-Beltrami spectrum, this project will develop a suite of novel tools for shape-based analysis of cortical, sub-cortical, and white matter structures. These tools will allow the robust and accurate comparison of structural variations in development and pathology across population.
TR&D 3 is tightly related to TR&D 1 and 2. For TR&D 1, this project will provide robust tools for the large scale study and organization of the shape representation of anatomical structures. The intrinsic embedding provides a natural way of organizing the Big Data of shapes. For TR&D 2, this project will provide shape-based tools for the fiber bundles produced that it will produce.
- Aim 1. To improve the conformal metric optimization on surfaces (CMOS) method for intrinsic shape analysis.
- Aim 2. To develop a population-based surface mapping method in the Laplace-Beltrami embedding space.
- Aim 3. To develop intrinsic surface analysis methods for white matter structures
- Aim 4. To apply intrinsic shape analysis tools to brain mapping research.
- Provide a complete solution from surface reconstruction, mapping, and quantification with intrinsic geometry.
- Provide a general solution of shape analysis of various anatomical structures.
- Develop a population-based method for large scale shape modeling
- Enables shape-based analysis of white matter fiber bundles and connectivity evaluation