Information Discovery on Electronic Medical Records

As the use of Electronic Medical Records (EMRs) becomes more widespread, so does the need to search and provide effective information discovery on them. Information discovery methods will allow practitioners and other healthcare stakeholders to locate relevant pieces of information in the growing corpus of available EMRs. The success of Web search engines has shown that keyword queries are a useful tool for locating relevant information in an intuitive and effective manner. However, questions arise of the form: What are the semantics of keyword queries on EMRs? What is a meaningful result? What is the role of medical and clinical ontologies and dictionaries like SNOMED (Systematized Nomenclature of Human and Veterinary Medicine) in answering such queries? In this position paper we introduce the problem of keyword-based information discovery on EMRs and enumerate the salient challenges that must be addressed to facilitate quality information discovery. The objective is to create interest in new medical information management research initiatives, and potentially create new paradigms for using medical data. The primary focus of the paper is the newest XML-based EMR standard created by the Health Level Seven (HL7) group, the Clinical Document Architecture (CDA) Release 2.0, although the same issues arise for any other standard hierarchical format.