På svenska

Presentation Information     2013-09-09 (14:15)   •  The seminar room at Vi2

Speaker Punam Saha
Title Fuzzy Digital Topology and Geometry in Medical Imaging Skeletonization and Other Approaches
Abstract The primary end-goal of most medical imaging research program is to collect information of internal human organs or tissues through a variety of in vivo or ex vivo imaging techniques. Often, medical imaging techniques suffer from limited spatial and temporal resolution, noise, backgroundinhomogeneity, and other artifacts leading to fuzzy representation of target objects in acquired images. Digital topology and geometry play important roles in medical image processing either by expanding the scope of target information or by providing a strong theoretical foundation to a process enhancing its stability, fidelity, and efficiency. The notions of digital topology and geometry are often intertwined in medical imaging applications and sometime it is difficult to draw a dividing line between them. During my talk, first, I will present a new framework for fuzzy skeletonization and its applications. Finally, I will briefly mention about other research works from my laboratory and other collaborative research groups which are related to fuzzy digital topology and geometry. Skeletonization provides a simple yet compact representation of an object and is widely used in medical imaging applications including volumetric, structural, and topological analyses, object representation, stenoses detection, path-finding etc. Literature of three-dimensional skeletonization is quite matured for binary objects. However, the challenges of skeletonization for fuzzy objects are mostly unanswered. Here, a framework and an algorithm for fuzzy surface skeletonization are developed using a notion of fuzzy grassfire propagation. Several concepts including fuzzy axial voxels, local and global significance factors are introduced. A skeletal noise pruning algorithm using global significance factors of individual branches is developed.