Research On Lung CT Image Segmentation Based On Computational Intelligence And Mathematical Morphology

Medical image processing is a technology which combine computer graphics and medical knowledge,its ultimate goal is to digitize medical image,allowing doctors`vision can perceive the image clearly,so as to distinguish the lesions.Medical image segmentation is the first step in medical image process,and is also the basis for subsequent image recognition and image understanding.Image segmentation is to divide the image into several different areas,extracting the interesting objects such as lesion part to do further research,and making the segmentation results close to the anatomical structures as far as possible,which provide a reliable basis for clinical diagnosis and pathology research.However,until now,with the development of image segmentation technology,there are still exist a variety of problems,such as the accuracy of segmentation and the degree of automation can`t meet the actual needs,So it still has great significance for the study of medical image segmentation methods.To solve these problems,this paper proposes a new framework for segmentation algorithm which adopts two novel computational intelligence and mathematical morphology computer system,Specific applications is the combination of the group search optimization algorithm(Group Search Optimiser, referred GSO) and improved mark control watershed transformation(Mark Controled Watershed Transformation, referred MCWT).The main idea of the algorithm framework:first,to do a series of image pre-processing,extracting the target object,then labeling the image,using the improved watershed transform algorithm to segment the labeled image to obtain the segmentation result.Finally,utilizing GSO optimization algorithm to optimize the segmentation results repeatedly according to some certain fitness function to produce better segmentation results automaticaly.In this paper, the algorithm applies on the latest CT imaging equipment, Carbon Nanotubes( referred CNT) CT, through the experiment simulation of the lung CT image segmentation,which demonstrates that this new algorithm has robust automatic and semi-automatic segmentation and more accuracy results, and also has great significance for the key algorithms in high-end medical imaging equipment.