Mining Electronic Health Records to Guide and Support Clinical Decision Support Systems

Abstract

Clinical decision support systems require well-designed electronic health record (EHR) systems and vice versa. The data stored or captured in EHRs are diverse and include demographics, billing, medications, and laboratory reports; and can be categorized as structured, semi-structured and unstructured data. Various data and text mining techniques have been used to extract these data from EHRs for use in decision support, quality improvement and research. Mining EHRs has been used to identify cohorts, correlated phenotypes in genome-wide association studies, disease correlations and risk factors, drug-drug interactions, and to improve health services. However, mining EHR data is a challenge with many issues and barriers. The aim of this chapter is to discuss how data and text mining techniques may guide and support the building of improved clinical decision support systems.