An EHR Study Based On Ontology

In recent years, ontology is been concerned in knowledge engineering, artificial intelligence, Semantic Web and related areas, and ontology also has been widely used to resolve the issues, which are about the knowledge reuse and sharing, knowledge acquisition and systems integration. In the medical field, some features of domain knowledge,such as the strong subjectivity of domain knowledge, uncertainty, ambiguity and controversial, make building a expertise knowledgebase very difficult. Considering this situation, we use ontology theory to study of EHR(electronic health records),try to build an EHR knowledge base, and on this basis, wo also attempt to generate rules for EHR knowledge base and reasoning by the rules. At last, we present the clinical testing model.This paper is divided into five parts, in the first part, we analyze the background to this study. Ontology have been rapid development in recent years, the Ministry of Health carries out the national standards of EHR. After introducing the recent situation about Ontology, Unified Medical Language System and HER, this paper presents a goal of building an ontology model of HER, and plans to take advantage of this ontology model to build the knowledge base of EHR, research rule-based reasoning on this knowledge base.The second part of this article, we introduce the national standards for EHR and the ontology theory. Finally, we expound the rule-generated methods for the knowledge base.The third part of this paper, we study the national standards for EHR, using the ontology theory,ontology-building tools,modeling languages and related methods, we successfully completed the building of EHR ontology model. Before the building process starting, we present the application framework for this HER ontology model, and underline two significant points about this study.This fourth part, while we use the ontology model to build an EHR knowledge base, we found the issue of standardizing terminology. So, we proposed a terminology extraction algorithm and the negative word selection algorithm.This fifth part, we study the theory of Rough-Set and Decision-Tree, present the rule-generated algorithm, and give examples to validate the algorithm.while using EHR knowledge base to diagnose and reasoning, this paper presents a clinical testing model.Finally, this paper reviews the studied issues, and prospects the future development of EHR.