Segmentation using hybrid discriminative generative label fusion of multiple atlases

A method for segmenting a target image comprising:

• receiving the target image of an anatomical structure, wherein the target image is stored in a memory;
• registering a plurality of atlases to the target image, each of the atlases including an image and a plurality of labels corresponding to portions of the image;
• selecting a plurality of registered atlases;
• transferring the labels of selected registered atlases to the target image;
• combining the labels that are transferred to the target image using a fusion of a discriminative model and a generative model to obtain fused labels for the target image;
• processing, by a processor, the target image using the fused labels to obtain a segmentation of the target image isolating the anatomical structure;
• and
• outputting, to a display, the segmentation of the target image isolating the anatomical structure, wherein the segmentation of the target image is displayed.