Kozaki Kouji, Yamagata Yuki, Mizoguchi Riichiro, Imai Takeshi, Ohe Kazuhiko
The Institute of Scientific and Industrial Research, Osaka University, 8-1 Mihogaoka, Ibaraki, Osaka, 567-0047, Japan.
National Institute of Biomedical Innovation, Health and Nutrition, 7-6-8, Saito-Asagi, Ibaraki, Osaka, 567-0085, Japan.
J Biomed Semantics. 2017 Jun 19;8(1):22. doi: 10.1186/s13326-017-0132-2.
Medical ontologies are expected to contribute to the effective use of medical information resources that store considerable amount of data. In this study, we focused on disease ontology because the complicated mechanisms of diseases are related to concepts across various medical domains. The authors developed a River Flow Model (RFM) of diseases, which captures diseases as the causal chains of abnormal states. It represents causes of diseases, disease progression, and downstream consequences of diseases, which is compliant with the intuition of medical experts. In this paper, we discuss a fact repository for causal chains of disease based on the disease ontology. It could be a valuable knowledge base for advanced medical information systems.
We developed the fact repository for causal chains of diseases based on our disease ontology and abnormality ontology. This section summarizes these two ontologies. It is developed as linked data so that information scientists can access it using SPARQL queries through an Resource Description Framework (RDF) model for causal chain of diseases.
We designed the RDF model as an implementation of the RFM for the fact repository based on the ontological definitions of the RFM. 1554 diseases and 7080 abnormal states in six major clinical areas, which are extracted from the disease ontology, are published as linked data (RDF) with SPARQL endpoint (accessible API). Furthermore, the authors developed Disease Compass, a navigation system for disease knowledge. Disease Compass can browse the causal chains of a disease and obtain related information, including abnormal states, through two web services that provide general information from linked data, such as DBpedia, and 3D anatomical images.
Disease Compass can provide a complete picture of disease-associated processes in such a way that fits with a clinician's understanding of diseases. Therefore, it supports user exploration of disease knowledge with access to pertinent information from a variety of sources.
医学本体有望促进对存储大量数据的医学信息资源的有效利用。在本研究中，我们聚焦于疾病本体，因为疾病的复杂机制与各个医学领域的概念相关。作者开发了一种疾病的河流流动模型（RFM），该模型将疾病捕获为异常状态的因果链。它表示疾病的病因、疾病进展以及疾病的下游后果，这与医学专家的直觉相符。在本文中，我们讨论了基于疾病本体的疾病因果链事实库。它可能成为先进医学信息系统的宝贵知识库。
我们基于疾病本体和异常本体开发了疾病因果链事实库。本节总结这两种本体。它被开发为链接数据，以便信息科学家可以通过用于疾病因果链的资源描述框架（RDF）模型使用SPARQL查询来访问它。
我们根据RFM的本体定义设计了RDF模型，作为事实库的RFM实现。从疾病本体中提取的六个主要临床领域的1554种疾病和7080种异常状态，作为带有SPARQL端点（可访问API）的链接数据（RDF）发布。此外，作者开发了疾病指南针，这是一种疾病知识导航系统。疾病指南针可以浏览疾病的因果链，并通过两个网络服务获取相关信息，包括异常状态，这两个网络服务从链接数据（如DBpedia）提供一般信息以及3D解剖图像。
疾病指南针可以以符合临床医生对疾病理解的方式提供与疾病相关过程的全貌。因此，它支持用户探索疾病知识，并从各种来源获取相关信息。
