One of the most important benefits of implementing SNOMED CT is the great power it provides to data analytics. This section highlights a set of tools on how data analytics processes can benefit from the use of SNOMED CT and the capabilities provided by the use of an ontology.

Data Analytics using SNOMED CT can be used in different scenarios, and applying different techniques:

1. Scenarios for Data Analytics
   1. Point of care, supporting clinician activities, organizing information, etc.
   2. Population-based analytics, statistical reports, surveillance, etc.
   3. Clinical research, specific patterns, research candidates, outcomes, etc.
2. Data Analytics Techniques
   1. Subsets, hierarchies
   2. Description Logic expressions
   3. ECL - Expressions Constraints Language queries
   4. Mapping to statistical classifications

Implementing analytics in these scenarios requires defining a process that starts with obtaining sets of SNOMED CT ConceptIds using one of the techniques described, and then matching with fields in information stored in clinical repositories.

Powering Data Analytics with SNOMED CT

SNOMED International has created a demonstration tool for Data Analytics, based on synthetic data to avoid any privacy risks. This tool is used in seminars and workshops to demonstrate the capabilities of SNOMED CT for analytics connected to a virtual Electronic Health Records system.

The tools are open-source and are available at: https://github.com/IHTSDO/health-data-analytics

In this video the tool is demonstrated live:

ECL (Expression Constraint Language) Implementations

The Expression Constraint Language is a fundamental component of the data analytics process, converting clinical requirements into computable queries that select a set of concepts with a specific meaning. Some examples of implementations, which support the SNOMED CT Expression Constraint Language, are listed below.