Neural Integration Approach for Subcortical Structure Segmentation

A neural approach is presented to automatically segment subcortical brain structures from MR image. Spatial information from anatomy and tissue characteristics of the brain structures is employed in the method. The spatial locations of the structures are learned by an improved fuzzy neural approach. The decision of a particular voxel belonging to a particular structure, rendered by spatial and tissue characteristics information are combined by a fuzzy fusion algorithm. Experiments on an ensemble of MR brain scans show promise of the current approach