Refinement of object-based segmentation Public Deposited

Downloadable Content

Download PDF
Last Modified
  • March 21, 2019
  • Levy, Joshua Howard
    • Affiliation: College of Arts and Sciences, Department of Computer Science
  • Automated object-based segmentation methods calculate the shape and pose of anatomical structures of interest. These methods require modeling both the geometry and object-relative image intensity patterns of target structures. Many object-based segmentation methods minimize a non-convex function and risk failure due to convergence to a local minimum. This dissertation presents three refinements to existing object-based segmentation methods. The first refinement mitigates the risk of local minima by initializing the segmentation closely to the correct answer. The initialization searches pose- and shape-spaces for the object that best matches user specified points on three designated image slices. Thus-initialized m-rep based segmentations of the bladder from CT are frequently better than segmentations reported elsewhere. The second refinement is a statistical test on object-relative intensity patterns that allows estimation of the local credibility of a segmentation. This test effectively identifies regions with local segmentation errors in m-rep based segmentations of the bladder and prostate from CT. The third refinement is a method for shape interpolation that is based on changes in the position and orientation of samples and that tends to be more shape-preserving than a competing linear method. This interpolation can be used with dynamic structures and to understand changes between segmentations of an object in atlas and target images. Together, these refinements aid in the segmentation of a dense collection of targets via a hybrid of object-based and atlas-based methods. The first refinement increases the probability of successful object-based segmentations of the subset of targets for which such methods are appropriate, the second increases the user's confidence that those object-based segmentations are correct, and the third is used to transfer the object-based segmentations to an atlas-based method that will be used to segment the remainder of the targets.
Date of publication
Resource type
Rights statement
  • In Copyright
  • Pizer, Stephen M.
  • Open access

This work has no parents.