Segmentation of multispectral MR images using a hierarchical self-organizing map

The application of a hierarchical self-organizing map (HSOM) to the problem, of segmentation of multispectral magnetic resonance (MR) images is investigated. The HSOM is composed of several layers of the self-organizing map (SOM) organized in a pyramidal fashion. The SOM has been used for the segmentation of multispectral MR images but the results often suffer from undersegmentation and oversegmentation. By combining the concepts of self-organization and topographic mapping with multiscale image segmentation, the HSOM is seen to overcome the major drawbacks of the SOM. The segmentation results of the HSOM are compared with those of the SOM and the k-means clustering algorithm on multispectral MR images of the human brain representing both, normal conditions and pathological conditions such as multiple sclerosis. The multiscale segmentation results of the HSOM are shown to have interesting consequences from the viewpoint of clinical diagnosis of pathological conditions.