Home
Organizers
Invited Speakers
Schedule & Material
Abstracts
Related Publications & Code
Contact Us

 

Shape models have attracted the attention of many researchers in the past few years due to their successful role in many medical image segmentation tasks. Many of the most successful automatic segmentation algorithms evaluated during the previous MICCAI Grand Challenge competitions made effective use of shape models to provide prior knowledge for these various segmentation tasks. An increasingly wider range of methodologies have emerged for the construction, representation, and application of these shape models. Often, the shape models are only briefly presented in the context of specific segmentation algorithms. A thorough, detailed discussion focusing upon shape model construction and its uses is sometimes overlooked. The goal of this tutorial is to address this topic and to provide a unified framework for in-depth discussion of the construction and application of shape models in medical image segmentation.

The tutorial will review different methodologies for the use of shape models in medical image segmentation algorithms. The main focus will be on describing different approaches to building shape models that are used as prior knowledge for image segmentation algorithms and the different techniques to use these models for segmentation. The tutorial will consist of three main themes: 1) Recent clinical applications that require segmentation as a pre-requisite, 2) Mesh and Image based atlas construction methodologies, and; 3) Integration of shape-models into image segmentation algorithms.