基于本体的医学图像分析研究

医学图像在临床诊断与治疗中的应用日益广泛，如何有效利用影像管理系统中的大量图像来辅助医生进行分析与诊断，是一个非常重要的问题。传统的基于文本关键字的图像检索方法已不能满足对大型医学图像数据库检索的需要，在分析传统医学图像检索方法不足的基础上，结合模糊本体概念提出一种基于模糊本体的医学图像分析方法。该方法利用模糊本体中隶属度的概念，对查询语句和图像关键词向量空间模型进行模糊本体概念映射和概念相似度计算。实验证明，这一方法具有良好的性能。

The Research of Medical Image Analysis Based on Ontology

Medical Image has been increasingly applied in clinical diagnosis and treatment, it is very important to make use of large numbers of images in medical image management system in order to help clinician to analyze and diagnose. The traditional information retrieval techniques are not fit for retrieving large scale medical image databases. It is a very promising idea to introduce Ontology-based Image Retrieval technique into indexing medical image databases. Based on the deficiency of traditional medical image method and the concept of fuzzy ontology, we present a novel medical image method. In this method, concept mapping and similarity calculating are made based on the membership-degree concept of fuzzy ontology, which make it possible for us to analyze image on concept level. Finally experiments are carried out to test it performance, and the result shows its efficiency.