Thesis Defense: Segmentation of Medical Images Under Topological Constraints
Speaker: Florent Segonne , MIT CSAILRelevant URL:
Date: September 27 2005
Time: 10:00AM to 11:30AM
Location: Patil/Kiva, 32-G449
Host: Eric Grimson, Acting Head, EECS
Segmentation of Medical Images Under Topological Constraints
Major advances in the field of medical imaging over the past two decades have provided physicians with powerful, non-invasive techniques to probe the structure, function, and pathology of the human body. This increasingly vast and detailed amount of information constitutes a great challenge for the medical imaging community, and requires significant innovations in all aspect of image processing.
The accurate and topologically correct delineation of anatomical structures from medical images is a critical step for many clinical and research applications. In this thesis, we extend the theoritical tools applicable to the segmentation of images under topology control, apply these new concepts to broaden the class of segmentation methodologies, and develop generally applicable and well-founded algorithms to achieve accurate segmentations of medical images under topological constraints.
First, we introduce a novel digital concept, which offers more flexibility in controlling the topology of digital segmentations. Second, we design a novel level set framework that provides a subtle control over the topology of the level set. Our method constitutes a trade-off between traditional level sets and topology preserving level sets. Third, we develop an algorithm for the retrospective topology correction of 3D digital segmentations. Our method is phrased within the theory of Bayesian parameter estimation, and integrates statistical information into the topology correction process. In addition, no assumption is made about the topology of the initial input images. Finally, we propose a genetic algorithm to accurately correct the spherical topology of cortical surfaces. Unlike existing approaches, our method is able to generate several potential topological corrections and to select the maximum-a-posteriori retessellation in a Bayesian framework. Our approach integrates statistical, geometrical and shape information into the correction process, providing optimal solutions relatively to the MRI intensity profile and the expected curvature.
See other events that are part of
See other events happening in September 2005