FIELD: physics.
SUBSTANCE: system comprises: workstation which segments a volume of interest in previously generated diagnostic images; a processor is programmed to register the segmented previously generated images and merge the segmented previously generated images into a probability map which depicts the probability that each voxel represents the volume of interest or background, and a mean segmentation boundary; a segmentation processor registers the probability map with a current diagnostic image to generate a transformed probability map; a previously-trained classifier classifies voxels of the diagnostic image using the probability that each voxel depicts the volume of interest or background; a merge processor merges the probabilities from the classifier and the transformed probability map; a segmentation boundary processor determines the segmentation boundary for the volume of interest based on the current image based on the merge probabilities.
EFFECT: high accuracy of segmentation of diagnostic images.
13 cl, 5 dwg
