It has been recently shown that thalamic nuclei can be automatically segmented using diffusion tensor images (DTI)
under the assumption that principal fiber orientation is similar within a given nucleus and distinct between adjacent
nuclei. Validation of these methods, however, is challenging because manual delineation is hard to carry out due to the
lack of images showing contrast between the nuclei. In this paper, we present a novel gray-scale contrast for DTI
visualization that accentuates voxels in which the orientations of the principal eigenvectors are changing, thus providing
an edge map for the delineation of some thalamic nuclei. The method uses the principal fiber orientation computed from
the diffusion tensors computed at each voxel. The three-dimensional orientations of the principal eigenvectors are
represented as five dimensional vectors and the spatial gradient (matrix) of these vectors provide information about
spatial changes in tensor orientation. In particular, an edge map is created by computing the Frobenius norm of this
gradient matrix. We show that this process reveals distinct edges between large nuclei in the thalamus, thereby making
manual delineation of the thalamic nuclei possible. We briefly describe a protocol for the manual delineation of thalamic
nuclei based on this edge map used in conjunction with a registered T1-weighted MR image, and present a preliminary
multi-rater evaluation of the volumes of thalamic nuclei in several subjects.
Purpose: By incorporating high-level shape priors, atlas-based segmentation has achieved tremendous success
in the area of medical image analysis. However, the effect of various kinds of atlases, e.g., average shape model,
example-based multi-atlas, has not been fully explored. In this study, we aim to generate different atlases and
compare their performance in segmentation.
Methods: We compare segmentation performance using parametric deformable model with four different atlases,
including 1) a single atlas, i.e., average shape model (SAS); 2) example-based multi-atlas (EMA); 3) cluster-based
average shape models (CAS); 4) cluster-based statistical shape models (average shape + principal shape variation
modes)(CSS). CAS and CSS are novel atlases constructed by shape clustering. For comparison purpose, we also
use PDM without atlas (NOA) as a benchmark method.
Experiments: The experiment is carried on liver segmentation from whole-body CT images. Atlases are
constructed by 39 manually delineated liver surfaces. 11 CT scans with ground truth are used as testing data
set. Segmentation accuracy using different atlases are compared.
Conclusion: Compared with segmentation without atlas, all of the four atlas-based image segmentation methods
achieve better results. Multi-atlas based segmentation behaves better than single-atlas based segmentation. CAS
exhibit superior performance to all other methods.