摘要
医学类知识图谱是现代医疗的基础,阻生智齿是一种常见疾病,发病率很高,但目前的研究还缺乏系统性和全面性。本文利用自然语言处理技术,参考口腔医学领域权威著作、医学百科、真实病例等资源,用BERT-BiLSTM-CRF模型进行实体抽取,从而构建阻生智齿知识图谱,包含阻生智齿的形态分类、治疗方案、治疗药物等相关信息,为大众提供阻生智齿科普性平台,为专业医疗人员提供系统的病例整合平台,推动阻生智齿领域的研究。
Medical knowledge graph is the foundation of modern medical care.Impacted wisdom teeth,as a common disease,have a high incidence.However,current research is lack of systematic and comprehensive.In this paper,using natural language processing technology,referring to authoritative works in the field of stomatology,medical encyclopedia,real cases and other resources,the BERT-BiLSTM-CRF model is used to extract entities,so as to construct a knowledge graph of impacted wisdom teeth,including morphological classification,treatment plan,treatment drugs and other related information of impacted wisdom teeth.It will provide a popular science platform for the public,and a systematic case integration platform for professional medical personnel to promote research in the field of impacted wisdom teeth.
作者
郑宇辰
段旭博
杨威
Zheng Yuchen;Duan Xubo;Yang Wei(Medical College,Guizhou University,Guiyang,Guizhou 550025,China)
出处
《计算机时代》
2022年第11期109-112,共4页
Computer Era
基金
黔科合支撑([2022]一般272)。
