Automatic Diagnosis Of COVID-19 Infection Based On Ontology Reasoning

COVID-19 has spread globally and has become the focus of global attention,and due to its rapid transmission speed and high contagiousness,the rapid screening and isolation of suspected and confirmed cases is an effective means to reduce the rate of COVID-19 infection.Due to the sudden outbreak of the epidemic,the research and application of COVID-19 ontology at home and abroad is still in the preliminary stage.In this context,this paper use artificial intelligence technology to build a new crown pneumonia diagnosis ontology to assist medical diagnosis,which can optimize the diagnostic workflow and reduce the burden on medical staff,which is particularly important for controlling the spread of the virus and improving the level of COVID-19 treatment.In this paper,we first constructed the COVID-19 diagnosis ontology against the background of the COVID-19,and further built the COVID-19 automatic diagnosis system.The specific research content is as follows:1)With the help of Protégé,the “seven-step method” was used to construct the COVID-19 diagnosis ontology from top to bottom,including diagnostic rules for suspected and confirmed cases translated from Chinese government documents;in addition,SWRL rules have been added to infer the close contact relationship of the population,thereby assisting the effective diagnosis of COVID-19;and finally,real case data was collected from the Ningbo Municipal Health Commission for verification.Experimental results prove that the COVID-19 diagnosis ontology shows good performance in diagnosing suspected and confirmed cases.2)In order to facilitate and efficient use,we have built an automatic diagnosis system for COVID-19 using Python and Django,which can preprocess the unstructured data entered by users,and the structured data obtained will automatically diagnose suspected and confirmed cases and output the results after matching the diagnostic rules of COVID-19,and the patient’s structured information will be saved to the database and displayed on the interface.The system reduces the manual input of the COVID-19 diagnostic process to a certain extent,which can effectively alleviate the pressure on healthcare workers during the epidemic and further reduce the infection rate and fatality rate of COVID-19.