文章基本信息

标题：Semantically Enhanced Frequent Events Mining in Electronic Health Records
其他标题：Semantically Enhanced Frequent Events Mining in Electronic Health Records
本地全文：下载
作者：Svetla Boytcheva
期刊名称：Brain. Broad Research in Artificial Intelligence and Neuroscience
印刷版ISSN：2067-3957
出版年度：2019
卷号：10
期号：1
页码：43-54
出版社：EduSoft publishing
摘要：This paper proposes context based approach for frequent events mining (FEM) in Electronic Health Records (EHR). The majority of FEM methods do not take in consideration the context information of the analyzed dada. EHRs contain rich context information like demographic data, encounters, vital parameters, diagnoses, lab tests values, and prescribed therapy. Such information is crucial for proper interpretation of the complex temporal clinical events. Some applications in comorbidity identification, risk factors analysis and patients phenotyping are presented to illustrate the proposed method. Experiments were run on large collections of pseudoanonimized reimbursement requests submitted to the Bulgarian National Health Insurance Fund in 2010-2016 for more than 5 million citizens yearly. Effective explication of comorbidities and characterization of risk factors can fill knowledge gaps and assist informed clinical decision-making.
关键词：Frequent Patterns Mining;Data Mining; Knowledge Discovery;Health Informatics