Research On Clinical Case Knowledge Base Based On Ontology

In the medical field，currently, the case is extremely a valuable source ofknowledge. It would be an important part of clinical decision support system(CDSS).The case-based reasoning technology, which makes up for the shortage of rule-basedreasoning technology that has the difficulty of acquiring rules and weak ability toexplain anomalies, is widely used on CDSS. Based on the idea of ontology andanalysis of clinical medical features, this paper proposes the acquisition,representation, reasoning and the construction method of clinical case knowledge. Themain research contents are listed as follows:1. According to the detailed analysis of clinical medical features, this paper putsup the representation of clinical medical ontology. Based on the analysis offormal representation of knowledge and the model representation of clinical medicalknowledge, this paper proposes the medical ontology-based representation of clinicalcase knowledge.2. Based on the analysis of research on existing medical ontology at home andbroad, this paper proposes the construction method of clinical medical ontology,especially of the concept dictionary of medical ontology. Further on, with the study ofcase base reasoning, the paper proposes the representation of clinical medical case.Also the construction of the clinical case knowledge base is given.3. The paper analyzes the information extraction of text clinical case file, and theadvantages and disadvantages of methods of medical name entity recognition. On thestudy of methods of structured and semi-structured data acquisition, the standardprocess and methods of clinical medical case knowledge acquisition is given, alongwith the XML-described representation of a clinical case knowledge base. 4. Similarity calculation is the most critical problem of case-based reasoning.This paper designs the similarity calculation methods in connection with the variety ofdata types, such as text type, degree type, value type, date time type and so on,through analyzing the existing method on similarity calculation. The methods oncalculation of similarity weight are put forward, which are tested and verified throughexperimentation.