Clustering Ensemble Technique Applied in the Discovery and Diagnosis of Brain Lesions

Medical image based computer aided diagnosis is considers to be an important and challenging task, it has extracted more and more research work in recent years. Due to its interdisciplinarity and complexity, there remain many problems not solved. In this paper, a novel diagnosis method named SeCED is proposed, which utilized as the core mechanism of our medical image based computer aided encephalopathy diagnosis system. The SeCED is built on a two-level architecture, where the kM-DBSCAN algorithm is employ as the base clusterer in each level and the k-Medoids algorithm is utilized to select a subset of clusterer for ensemble. Benefit from its selective clusterer ensemble technique, SeCED hold an improved generalization ability and achieved a satisfactory result of identify brain lesions in the real data experiment, and all the detailed experimental data will be presented in the end of this paper