摘要
利用SPO语义挖掘和实体识别技术提取文献中具有共病关系的疾病实体,在此基础上构建共病网络,运用链路预测方法发现潜在共病组合。糖尿病领域实证结果表明模型具有良好的有效性和准确性,能够对共病发病机制、疾病预防、临床诊疗等方面起到辅助作用。
SPO semantic mining and entity recognition technology are used to extract the disease entities with comorbidity relationship from literature.On this basis,the comorbidity network is constructed,and the link prediction method is used to predict the potential comorbidity combination.The empirical results in the field of diabetes show that the model has good effectiveness and accuracy,and can provide references for the pathogenesis,disease prevention,clinical diagnosis and treatment of comorbidity.
作者
关陟昊
单治易
林紫洛
杨雪梅
唐小利
GUAN Zhihao;SHAN Zhiyi;LIN Ziluo;YANG Xuemei;TANG Xiaoli(Institute of Medical Information,Chinese Academy of Medical Sciences&Peking Union Medical College,Beijing 100020,China)
出处
《医学信息学杂志》
CAS
2022年第4期27-32,共6页
Journal of Medical Informatics
基金
中国医学科学院医学与健康科技创新工程2021年重大协同创新项目“生物医学文献信息保障与集成服务平台”(项目编号:2021-I2M-1-033)。
关键词
共病
链路预测
SPO语义挖掘
comorbidity
link prediction
SPO semantic mining
