Biomedical Engineering Reference
In-Depth Information
volume data. The level set segmentation method, which is well documented in
the literature [5-8], creates a new volume from the input data by solving an initial
value partial differential equation (PDE) with user-defined feature-extracting
terms. Given the local/global nature of these terms, proper initialization of the
level set algorithm is extremely important. Thus, level set deformations alone are
not sufficient, they must be combined with powerful initialization techniques in
order to produce successful segmentations. Our level set segmentation approach
consists of defining a set of suitable preprocessing techniques for initialization
and selecting/tuning different feature-extracting terms in the level set algorithm.
We demonstrate that combining several preprocessing steps, data analysis and
level set deformations produce a powerful toolkit that can be applied, under the
guidance of a user, to segment a wide variety of volumetric data.
There are more sophisticated strategies for isolating meaningful 3D struc-
tures in volume data. Indeed, the so-called
segmentation problem
constitutes
a significant fraction of the literature in image processing, computer vision,
and medical image analysis. For instance, statistical approaches [9-12] typically
attempt to identify tissue types, voxel by voxel, using a collection of measure-
ments at each voxel. Such strategies are best suited to problems where the data
is inherently multivalued or where there is sufficient prior knowledge [13] about
the shape or intensity characteristics of the relevant anatomy. Alternatively,
anatomical structures can be isolated by grouping voxels based on local image
properties. Traditionally, image processing has relied on collections of edges,
i.e. high-contrast boundaries, to distinguish regions of different types [14-16].
Furthermore deformable models, incorporating different degrees of domain-
specific knowledge, can be
fitted
to the 3D input data [17, 18].
This chapter describes a level set segmentation framework, as well as the the
preprocessing and data analysis techniques needed to segment a diverse set of
biological volume datasets. Several standard volume processing algorithms have
been incorporated into framework for segmenting conventional datasets gener-
ated from MRI, CT, and TEM scans. A technique based on moving least-squares
has been developed for segmenting multiple nonuniform scans of a single object.
New scalar measures have been defined for extracting structures from diffusion
tensor MRI scans. Finally, a direct approach to the segmentation of incomplete
tomographic data using density parameter estimation is described. These tech-
niques, combined with level set surface deformations, allow us to segment many
different types of biological volume datasets.
