Page tree

Title: Semantic Query

Presenters: Balazs Banfai and Brandon Ulrich , B2i Healthcare


Anyone interested in using SNOMED CT’s semantics to search and analyse Electronic Health Records and clinical documents.


To understand techniques for leveraging SNOMED CT’s semantics in clinical document search and analysis.


The availability of semantically rich electronic health records utilizing SNOMED CT as a reference terminology continues to grow, providing new opportunities to improve patient care and reduce costs. However, traditional data warehouses struggle to unleash the full semantic meaning within the health records, as the data is built around a limited number of concepts.  

This presentation suggests an alternate strategy for executing meaningful queries. EHR data is represented using an information model bound to SNOMED CT terminology, where the information model is agnostic to the underlying standard (e.g. CIMI reference model, Singapore's Logical Reference Model, the UK's Logical Record Architecture, HL7, openEHR). Meaningful queries can be formulated using a query language that utilizes the SNOMED CT compositional grammar for post-coordinated expressions. This allows querying on not only the concept hierarchy but also the defining relationships as well, resulting in semantically aggregated patient data.   Complex queries can be executed in real-time for millions of EHRs without the need for extraction and aggregation to analytical stores.   The results of the query can be further analysed using a cloud-based analytics engine.

Additional materials

  1. B2i Healthcare