Research On Outlier Analysis With Medical Application

With the rapid development of modern information technology,a great deal of data has been accumulated in many fields.People expect to discover the knowledge and rules existing in these data,which just brings the study of data mining and the development of its technology.Data mining aims to extract the implicit,Previously unknown and potentially useful knowledge from voluminous,non-complete,fuzzy,stochastic data.As a comprehensive field of crossing mu1ti-subject,data mining involves many subjects such as database , statistic , machine learning , high performance computing,pattern recognition,neural network and data visualization etc.Cluster analysis is an important technology in data mining.Clustering processes are always carried out in the condition with no pre-known knowledge,so the mostresearch task is to solve that how to get the clustering result in this premises.The most research about clustering is focused on clustering algorithms,the main purpose is to produce practical algorithms with better performance.Up to now,many clustering algorithms have been presented,but these algorithms are only suited special problems and users.Furthermore,they are imperfect both theoretically and methodologically, even severe fault.Optimizing deeply clustering algorithms will not only help to perfect its theory,but also its popularization and application.Image mining is an important and difficult topic.There is a lack of effective tools for image mining,mining process needs human interfere and cannot be completely automated.In hospital,doctors detect the medical images usually by their personal experience, but their personal experiences can not reach every aspect of a mater, and there are always many other stochastic things disturbing him,all of these may bring on the mistakes of his examines.So, it's very important to make the way of detecting illness to be standardization by the help of computer techniques.The problem of outlier analysis has been variously called outlier mining,anomaly detection exception mining,detecting rare events,mining rare classes,deviation detection,etc.Outlier may be"dirty data",but it also can means meaningful event corresponding to the reality.From the point of knowledge discovery,rare events are often more interesting and valuable than others in many domains,where the rare events'importance is quite high compared to other events,making their detection and analysis extremely important.In this paper, we analyse traditional data mining technology gaps in image mining, describe the character and the status of image mining , and in a comprehensive, in-depth grasp of data mining technology based on the combination of image processing and medical knowledge,we give a new pixel-based medical image clustering outlier analysis technique,describing the specific application of outlier analysis in CT images.We will describe the whole process of the outlier mining.Firstly,the outlier mining process extracts CT images features and preprocess the data, and then divides the images by the way of clustering method on image pixels,constructs the vectors of parameter, and mines the outliers. With the results,we can help doctors to detect illness by a more efficient way.