Entropy-Based Particle Correspondence
Dr. Ipek Oguz - UNC Chapel Hill
Thursday, 11 October 2012, 14:40-15:30 @FENS 2008
Statistical shape analysis of anatomical structures plays an important role in many medical image analysis applications such as understanding the structural changes in anatomy in various stages of growth or disease. In many circumstances, establishing accurate correspondence across object populations is essential for such statistical shape analysis studies. In this talk I will present an entropy-based correspondence framework for computing point-based correspondence among populations of surfaces in a group-wise manner. This robust framework is parameterization-free and computationally efficient. The core principles of this method will be reviewed, as well as various extensions to deal effectively with surfaces of complex geometry and application-driven correspondence metrics. Synthetic and biological datasets will be used to illustrate the concepts proposed and to compare the performance of this framework to existing correspondence techniques.
Dr. Oguz is an assistant professor with joint appointments in the Departments of Computer Science and Psychiatry at the University of North Carolina at Chapel Hill (UNC), where she leads the small animal imaging group. Her main field of research is medical image processing and analysis, with a particular focus on neuroimaging using MRI and diffusion tensor imaging. Research interests include surface correspondence, connectivity analysis, morphometry, shape analysis, atlas building and segmentation.