Application Of Clustering Analysis In Medical Image

Currently, medical digital image information systems are increasingly popular, the rapid development of medical imaging equipment, imaging resolution gradually improved, the image of the information contained in tissues and organs of patients with a richer, while the number of CT images, low contrast, it is difficult to distinguish between lesions and normal tissues and organs. Medical image data mining technology is an effective solution to medical image processing and analysis of one of the important issues, also researchers in recent years concern has been a hot issue.This paper reviews foreign and the domestic medical image mining technology research, medical image data object for the complexity and particularity of images of existing data mining algorithms to compare the screening and improvements in exploration and discovery for medical image data mining methods and algorithms. Theoretical and applied research from two aspects, to explore suitable for medical image data mining techniques, to reveal the normal human tissues and organs of the characteristics of image data and the rules and the relationship between the distribution of medical image data cluster analysis of its technology for human tissues and organs Automatic classification and automatic recognition of lesions images via open up new.The main innovation of research results as follows:In-depth study of the medical image data of the nuclear density function, data boxes and boxes strategy based on data similar to the method of kernel density configuration. In the medical image on the basis of density, with medical images based on similar amendments to the density structure of the clustering algorithm.