A Unique Method for 3D MRI Segmentation

MRI & X-ray are the two methodologies widely used to visualize human anatomy. X-ray imaging is used when the details of bones have to be visualized whereas magnetic resonance imaging (MRI) provides rich three dimensional (3D) information about the human soft tissue anatomy. Segmentation of MRI images using intensity values is severely limited owing to in-homogeneities and partial volume effects. Edge based segmentation methods suffer from unsharp edges and gaps in boundaries. A number of other methods are also used. Here a method is presented to segment multidimensional images using a multiscale stacking approach (hyperstack) with either single parent or multiparent linking. A hyperstack is a voxel-based multiscale data structure whose levels are constructed by convolving the original image with a Gaussian kernel of increasing width. Between voxels at adjacent scale levels, child-parent linkages are established according to a model-directed linkage scheme. In the resulting tree-like data structure, roots are formed to indicate the most plausible locations in scale space where segments in the original image are represented by a single voxel. The final segmentation is obtained by tracing all root linkages.This will facilitate segmentation of grey matter, white matter and cerebrospinal fluid with minimum user interaction. Multiscale linking gives a significantly improved segmentation as compared with conventional (single-parent) linking.