Digital Health Data Capture with a Controlled Natural Language.

Area of Science:

• Clinical Informatics
• Medical Informatics
• Natural Language Processing

Background:

• Health data has historically relied on written text, with the medical narrative fostering patient-clinician relationships.
• Natural language remains a time-tested, user-accepted technology for information exchange.
• Previous work introduced controlled natural language for semantic data capture at the point of care.

Purpose of the Study:

• To extend a controlled natural language system for capturing clinical data.
• To enable the incorporation of numerical measurement results and their associated units.
• To explore the integration of this enhanced system with emerging clinical information models.

Main Methods:

• Developed an extension to a controlled natural language interface.
• The system is linguistically interpreted from the Systematized Nomenclature of Medicine - Clinical Terms (SNOMED CT) conceptual model.
• Enabled capture of measurement results including numerical values and units.

Main Results:

• Successfully extended the controlled natural language system to include numerical data with units.
• Demonstrated a method for semantic data capture of measurements.
• Established a foundation for integrating with clinical information modeling.

Conclusions:

• Controlled natural language can be effectively extended to capture quantitative health data.
• This approach enhances semantic data capture at the point of care.
• The method shows potential for alignment with evolving clinical information modeling standards.