Construction And Application Of CVD Knowledge Graph Based On Neo4j Graph Database

With the growing of medical data,knowledge graph can link up fragmented data,use the existing data to mine valuable potential relationship,and realize the integration of medical knowledge.In the field of medical treatment,the diagnosis and treatment standards of chronic venous diseases are complex and the medical knowledge is professional,which leads to the asymmetry of medical information between doctors and patients.Aiming at the above problems,a method of constructing knowledge graph oriented to medical field is proposed,and the application of knowledge interrogation is realized based on the knowledge graph.Crawler crawls medical data,collects electronic medical records and clinical guidelines related to chronic venous disease,defines attribute diagrams of knowledge graphes,and imports Neo4j graph database via py2neo component and load csv command.In order to show the entity and relation of knowledge graph more intuitively,the visual query is separated by Vue Flask framework,and the query results are presented in the form of echarts graph.Knowledge question answering uses Aho-Corasick algorithm to classify and parse,and invokes corresponding Cypher query statements according to different classifications.At the same time,named entity recognition plays a key role in the construction of knowledge graph.BiLSTM-CRF model is applied to named entity recognition of medical texts.Based on the knowledge of chronic venous disease,knowledge graph can effectively organize and manage knowledge in the field of medical treatment,solve the problem of lack of knowledge graph in the diagnosis and treatment system of chronic venous disease,and provide a new idea for medical data processing.At the same time,it provides a visual medical platform for doctors and patients,which integrates relationship query,entity query,medical entity identification and knowledge question answering.