Kilic Ozgur, Dogac Asuman, Eichelberg Marco
Department of Computer Engineering, Software R&D Center, Middle East Technical University, Ankara 06531, Turkey.
IEEE Trans Inf Technol Biomed. 2010 May;14(3):846-53. doi: 10.1109/TITB.2010.2041029. Epub 2010 Apr 15.
Providing an interoperability infrastructure for Electronic Healthcare Records (EHRs) is on the agenda of many national and regional eHealth initiatives. Two important integration profiles have been specified for this purpose, namely, the "Integrating the Healthcare Enterprise (IHE) Cross-enterprise Document Sharing (XDS)" and the "IHE Cross Community Access (XCA)." IHE XDS describes how to share EHRs in a community of healthcare enterprises and IHE XCA describes how EHRs are shared across communities. However, the current version of the IHE XCA integration profile does not address some of the important challenges of cross-community exchange environments. The first challenge is scalability. If every community that joins the network needs to connect to every other community, i.e., a pure peer-to-peer network, this solution will not scale. Furthermore, each community may use a different coding vocabulary for the same metadata attribute, in which case, the target community cannot interpret the query involving such an attribute. Yet another important challenge is that each community may (and typically will) have a different patient identifier domain. Querying for the patient identifiers in the target community using patient demographic data may create patient privacy concerns. In this paper, we address each of these challenges and show how they can be handled effectively in a superpeer-based peer-to-peer architecture.
为电子健康记录(EHR)提供互操作性基础设施已被列入许多国家和地区电子健康计划的议程。为此指定了两个重要的集成规范,即“整合医疗企业(IHE)跨企业文档共享(XDS)”和“IHE跨社区访问(XCA)”。IHE XDS描述了如何在医疗企业社区中共享EHR,而IHE XCA描述了如何跨社区共享EHR。然而,IHE XCA集成规范的当前版本并未解决跨社区交换环境中的一些重要挑战。第一个挑战是可扩展性。如果加入网络的每个社区都需要与其他每个社区连接,即纯对等网络,这种解决方案将无法扩展。此外,每个社区可能对相同的元数据属性使用不同的编码词汇表,在这种情况下,目标社区无法解释涉及此类属性的查询。另一个重要挑战是每个社区可能(而且通常会)有不同的患者标识符域。使用患者人口统计数据在目标社区中查询患者标识符可能会引发患者隐私问题。在本文中,我们解决了这些挑战中的每一个,并展示了如何在基于超级对等体的对等架构中有效地处理它们。