Please use this identifier to cite or link to this item:
Title: 3D Segmentation of Soft Tissues by Flipping-free Mesh Deformation
Authors: DING FENG
Keywords: medical,image,segmentation,3D,deformable,model
Issue Date: 10-Dec-2010
Citation: DING FENG (2010-12-10). 3D Segmentation of Soft Tissues by Flipping-free Mesh Deformation. ScholarBank@NUS Repository.
Abstract: Medical image segmentation has been a very hot research topic over many years. In general, it is a highly challenging problem. Medical images usually have inhomogeneous voxel intensities. Boundaries of target objects may be indistinct in some regions. The shapes of the target objects can be very complex in 3D, and they may have large variance across different patients. Moreover, medical volume images usually contain 50 to 100 million voxels per data set, which is very challenging for a segmentation algorithm. Many existing segmentation algorithms are often plagued by the problems mentioned above. They tend to produce undesired segmentation results. Many of them resort to a globalshape constraint, which enable the segmentation result to resemble a normal shape in such low contrast regions. This strategy succeeds when the shapes of the target objects are regular, i.e., close to the normal shape. However, shapes of soft organs are highly variable across different patients. They are in general very difficult to be modelled statistically even with a large number of training samples because the shape variations have huge number of degrees of freedom. With limited number of training samples, they usually cannot achieve accurate results when segmenting such very different shapes. This thesis presents a novel approach to the segmentation of soft tissues in 3D volume images. The proposed approach uses a specially designed 3D quadrilateral mesh to explicitly represent and segment an object, which is much more efficient compared to voxel-based segmentation algorithms. Segmentation is achieved by evolving the mesh to register to the desired object boundary. The mesh evolution-based segmentation is significantly more efficient than volumetric approaches. The proposed algorithm does not require any shape constraints, and is flexible for segmenting target organs with large shape variations among patients. Test results on using the single-object segmentation algorithm to segment various abdominal organs show that the proposed algorithm achieved higher accuracy than other segmentation algorithms such as snake, level set and graph-cut in segmenting individual organs. It is also more time efficient. The proposed approach can be extended to segmenting multiple organs simultaneously. As the meshes for different organs constraint each other, the proposed approach is free from the over-segmentation problem. It has no leaking problem and is more noise resilient. Test results on the multiple-object segmentation algorithm demonstrate that it is able to segment multiple objects simultaneously and to improve the segmentation accuracy by overcoming the leakage problem that may happen in single-object segmentation.
Appears in Collections:Ph.D Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
dingfeng-thesis.pdf6.15 MBAdobe PDF



Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.