以本體論為基礎之糖尿病用藥推薦系統之研究

The Study of Anti-Diabetic Drugs Recommend System Based on Domain Ontology

黃勻厚(Yun-Hou Huang)

指導教授 ： 陳榮靜

朝陽科技大學/資訊學院/資訊管理系/碩士(2010年)

摘要

糖尿病是現今最常見的慢性疾病之一，依世界衛生組織預估，亞洲地區的糖尿病患者在2010年到2025年間，會成長約56%，且隨年齡增長，罹患率也隨之增加。由於現今醫藥學進步，因此越來越多降血糖藥物可供醫師來使用，所以使用推薦系統即時協助醫師，開立符合糖尿病患所需要的藥品是必需發展的技術。本研究將利用Protégé訂定糖尿病藥品規定的病患本體知識及相關知識，用於儲存與糖尿病用藥和病患檢測有關的本體論，利用SWRL建構糖尿病用藥之間的法則關係，並透過推理機制，將糖尿病用藥本體中的知識類別及概念的解釋，轉為推薦系統可以接受的格式。由於SWRL無法直接被運算，因此需透過XSLT進行格式轉換，最後將透過JESS進行推論，找出推薦符合檢測結果的糖尿病用藥，並顯示監測、症狀及副作用等實據。本研究主要將結合行政院衛生署台中醫院糖尿病專科醫師及美國臨床內分泌科醫師學會的糖尿病臨床治療指引，將藥品的性質、種類、配藥和副作用，做一個完善的整理與分析。並以Protégé 3.4建置本體知識，同時結合SWRL及JESS推論，先分析糖尿病的症狀找出相關藥品，於比較後選擇出最合適的用藥調配。

關鍵字

本體論 ； OWL ； SWRL ； 推薦系統 ； JESS

並列摘要

Diabetes mellitus is one of the most common chronic diseases in recent years. According to the World Health Organization, estimated diabetic patient numbers will increase 56 percent in Asia from the year 2010 to 2025. The number of anti-diabetic drugs that doctors are able to utilize increases as the development of pharmaceutical drugs continues. Construction of diabetes medication recommendation system for doctor is necessary. Our study utilized Protégé to expand upon the interrelated information between anti-diabetic drugs knowledge and patient ontology knowledge. Interrelated ontology stores diabetes medication and patient symptom information. We used SWRL to increase the diabetes medication association information. The knowledge classification and term-explanation can transform to a recommendation system acceptable language by reasoners. Due to the fact that SWRL cannot directly be used it needs to use XSLT to transfer SWRL to JESS acceptable language. After the system was able to transfer into JESS it was then able to recommended Diabetes Mellitus medication and produce information about symptoms, side effects, and ways of monitoring the disease. In this thesis, we synthesized a hospital specialist in Taichung’s Department of Health with the “American Association of Clinical Endocrinologists Medical Guidelines for Clinical Practice for the Management of Diabetes Mellitus”. It made for a perfect collation and analysis to build ontology knowledge about the drugs’ of nature attributes, type of dispensing and side effects. We utilized the mapping of different ontology to select out better strategies which combine the tools of SWRL and JESS. This system was able to analyze the symptoms of diabetes as well as compare related drugs to select the most appropriate drug.

並列關鍵字

JESS ； SWRL ； OWL ； Ontology ； Recommendation System

參考文獻

[1] M. Argüello and J. Des (2007), “Clinical practice guidelines: A case study of combining OWL-S, OWL, and SWRL,” Applications and Innovations in Intelligent Systems XV, pp. 19-32.

[2] T. Andreasen and H. Bulskov (2009), “Conceptual querying through ontologies,” Fuzzy Sets and Systems, Vol. 160, No. 15, pp. 2159-2172.

[3] D. Baorto, L. Li, and J. J. Cimino (2009), “Practical experience with the maintenance and auditing of a large medical ontology,” Journal of Biomedical Informatics, Vol. 42, No. 3, pp. 494-503.

[4] F. Bobillo, M. Delgado, J. Gómez-Romero, and E. López (2009), “A semantic fuzzy expert system for a fuzzy balanced scorecard,” Expert Systems with Applications, Vol. 36, No. 1, pp. 423-433.

[6] A. Coronato, M. Esposito, and G. D. Pietro (2009), “A multimodal semantic location service for intelligent environments: an application for Smart Hospitals,” Personal and Ubiquitous Computing, Vol. 13, No. 7, pp. 527-538.