Research And Application Of Medical Heterogeneous Data Integration Based On Ontology

The development of medical science is directly related to our life and health.With the development of big data,How to explore the future medical science with data innovation and quickly acquire information and enhance humanity in huge data resources is a real problem that needs to be explored.The construction of medical informatization has been continuously carried out,but the information systems of medical institutions have been relatively closed,and the medical and health data cannot be interconnected.Semantic interoperability refers to the ability of two or more systems or components to communicate better and exchange information,which ensures that heterogeneous systems use the same specifications to parse and process data,and ensures the medical data can be understood,resolved,and used unambiguously.The first step in realizing medical data analysis and semantic interoperability is the integration of medical data.In data integration,medical data is characterized by complex data types,large data vo'lumes,and heterogeneous data sources.It has brought great challenge to the development of medical informationization.Integrating massive medical heterogeneous data is an urgent problem to be solved in the process of promoting medical informationization.Based on the characteristics of medical data sources and the integration of heterogeneous data,we propose a technical solution to introduce semantic interoperability into medical data integration.This scheme uses medical ontology similarity detection algorithm in the data warehouse of medical field,which introduce the ontology building part to further solve the heterogeneous data integration problem.Our paper is divided into the following sections:1.Propose the ontology construction method in the medical field:Firstly,the method extracts structural information of different heterogeneous data sources to establish local ontology,then calculate similarity to perform ontology fusion.Finally removes multiple semantics through global ontology mapping data relationship,and guide the ETL process;2.Propose medical similarity SDAMO algorithm:In this paper,we propose a medical ontology similarity detection algorithm-SDAMO algorithm,which is suitable for processing medical data.The algorithm can play an important role in the ontology of the medical domain with large volume.Compared with the traditional similarity detection algorithm,this algorithm is more accurate and closer to actual needs.3.Propose medical text data integration program:For the integration problem of unstructured text data,we introduce a semantic integration scheme on text data by introducing SDAMO algorithm and introducing semantic information based on the mainstream text classification algorithm.Finally,the experiments prove that the above methods perform well in eliminating the semantic uncertainty of data in the medical field,and can improve the integration efficiency of heterogeneous data in the medical data warehouse,and it is practical to solve the problem of integration of medical heterogeneous data.