2004 Fiscal Year Annual Research Report
日英二ヵ国語医学オントロジーの構築と臨床情報システムへの応用と評価
Project/Area Number |
04F04234
|
Research Institution | Hokkaido University |
Principal Investigator |
櫻井 恒太郎 北海道大学, 大学院・医学研究科, 教授
|
Co-Investigator(Kenkyū-buntansha) |
JIANG Guoqian 北海道大学, 大学院・医学研究科, 外国人特別研究員
|
Keywords | Concept-oriented view / Knowledge-based system / Formal Concept Analysis / Information retrieval / Nursing terminology systems / Ontology / Knowledge representation / Clinical domains |
Research Abstract |
Concept-Oriented View Generation for Clinical Data Using Formal Concept Analysis In the study, the potential role of formal concept analysis in concept-oriented view generation for clinical data is investigated. A concept-oriented view generation system for clinical data was developed as a Protege-2000 JAVA tab plug-in. A collection of discharge summaries written in Japanese and represented by a XML file was mapped to Protege-2000 as a virtual patient record system. A natural language processing(NLP) module was developed in order to extract the relevant concepts from the virtual patient records, based on using a Japanese standard dictionary of diagnostic terms and procedural terms released by MEDIS-DC. And a formal concept analysis(FCA) module was developed to generate the concept oriented views for the clinical data. The system could generate the views not only for individual patient records using their structural knowledge, but also for similar cases. The related issues and the future works were discussed. Developing a Support Tool for Describing the Nursing Practice in Japan with ICNP Using Pretege-2000 The purpose of the study is to develop a support tool to facilitate the description of nursing practice in Japan with ICNP. A support system was developed, mainly by the following three steps: 1) representing the Nursing Master in Protege-2000; 2) representing the ICNP Japanese version in Protege-2000; 3) designing an ontology approach to support the description of nursing practice in Japan with ICNP. An algorithm was developed to automate the cross-mapping and a preliminary evaluation was performed to show its usefulness. The related issues and the future works were discussed.
|