Organs extraction using three-dimensional mathematical morphology

We propose a method to automate the extraction of organ areas from computing tomography (CT) images using three-dimensional mathematical morphology. Morphological segmentation provides us the theory and tools to analyze shapes directly. This characteristic leads itself well to analyzing and extracting organ areas according to size and grey level features. After a system for three-dimensional morphological processing was built, we employed a new approach, differential top-hats, to our new three-dimensional organ extraction system. The experimental results show that organ areas can be extracted automatically with a success rate of 96%