A novel tool for standardizing clinical data in a semantically rich model

Area of Science:

• Biomedical Informatics
• Semantic Web Technologies

Background:

• Standardizing clinical information is crucial for data interoperability and research quality.
• Semantic Web technologies, like Resource Description Framework (RDF), require semantically accurate data models.
• Developing such models and integrating disparate data sources presents significant challenges.

Purpose of the Study:

• To introduce the PennTURBO Semantic Engine, a tool for creating semantically rich, source-independent clinical data models.
• To demonstrate a method for programmatically defining and populating these models.

Main Methods:

• Utilizing ontologies as foundational building blocks for the data model.
• Transforming concise RDF data into a semantically rich model.
• Dynamically generating and executing SPARQL update statements for data transformation.
• Employing human-readable configuration files for model sharing and collaboration.

Main Results:

• The PennTURBO Semantic Engine successfully transforms RDF data into a semantically rich, source-independent model.
• The system defines relationships between class instances based on an application ontology.
• The engine supports generalizable data standardization across various data models and sources.

Conclusions:

• The PennTURBO Semantic Engine offers a robust solution for clinical data standardization.
• Its design promotes interoperability and facilitates high-quality clinical research.
• The tool's shareable configuration files encourage institutional collaboration on data models.