HIWAS: Enabling technology for analysis of clinical data in XML documents

The information contained in large collections of clinical data can be used for many valuable purposes, such as epidemiological studies, evidence-based medicine, monitoring compliance with best clinical practices, and cost-benefit analyses. However, the emerging standards for the electronic representation of clinical data, such as the Clinical Document Architecture (CDA) [4], are very complex and new tools are required to effectively extract and utilize the information contained in these documents. In this paper, we present HIWAS, a research prototype of a new tool that creates a structural summary of a collection of XML documents, thereby enabling users to find relevant information for a specific purpose within complex XML documents. A HIWAS user can create a target model that contains just the information they need, in a simplified representation that can be queried efficiently and is compatible with existing relational business intelligence technology. By making these complex XML documents digestible with conventional tools, HIWAS lowers a key barrier to meaningful use of aggregated clinical data. © 2011 VLDB Endowment.