| Background: At present,the number of diabetes patients at home and abroad is increasing,and the form of diabetes prevention and management is still very severe.The number of diabetic patients in China is the first in the world,and diabetes and its complications have brought a huge economic burden to the society,while the level of diabetes prevention and management at the grassroots level is still very inadequate.In order to help doctors improve their clinical diagnosis and treatment,the use of ontology techniques to integrate knowledge related to clinical diagnosis and treatment of diabetes and to provide clinical advice in combination with reasoning techniques has become a hot research topic.Purpose: Combined with the scenarios of grass-roots diabetes prevention and management applications,we analyze authoritative clinical guidelines at home and abroad,use ontology technology to build a diabetes ontology knowledge base for grassroots applications,and develop a diabetes assisted diagnosis and treatment system to provide standardized and effective diagnosis and treatment decision support for grassroots doctors to help implement the hierarchical diagnosis and treatment model and improve the status quo of diabetes management.Methods: Using the literature research method and content analysis method,the clinical guidelines,literature and policies related to diabetes management services in the field of diabetes were sorted out,and the construction of the ontology knowledge base in the field of diabetes was completed according to the seven-step ontology construction process by extending the four-dimensional universal human disease ontology model,referring to the Me SH vocabulary,the concept of diabetes in the DOID ontology,and some semantic relationships of UMLS.On this basis,the SWRL provided by the protégé 5.5.0 platform is used to construct diabetic semantic rules and realize ontological semantic inference.And the use of Jena reasoning machine,using Java programming language to build a diabetes assisted diagnosis and treatment system,to achieve diabetes assisted diagnosis and treatment function.Results: Based on the clinical guidelines and policy specifications of diabetes,the knowledge in the field of diabetes was manually sorted out,the ontology knowledge base of the diabetes field for the application of primary medical and health institutions was constructed,and the consistency test was carried out by using the Pellet reasoning machine.According to the relevant content of diabetic diseases in the knowledge base,five types of rules for diabetes diagnosis and classification,complication identification,treatment,referral and follow-up are compiled,and the semantic reasoning of the ontology model is realized on the protégéplatform.The constructed ontology model consists of 153 classes,97 individuals,11 object properties,49 data properties and 93 SWRL rules.Finally,we developed a diabetes assisted diagnosis and treatment system in Java environment,which combined with Jena inference engine to realize the personal information management function and assisted diagnosis and treatment function of diabetes patients.Conclusions: The diabetes ontology knowledge base constructed in this study combines the application scenarios of diabetes prevention and management in primary medical and health institutions,and designs an auxiliary diagnosis and treatment system for diabetes,which has certain practical significance,and can also provide a basis for the construction of disease ontology for the application of other chronic diseases in primary medical and health institutions. |
PDF Full Text Request