Combining infobuttons and semantic web rules for identifying patterns and delivering highly-personalized education...

Area of Science:

• Biomedical Informatics
• Clinical Decision Support
• Semantic Web Technologies

Background:

• Infobuttons are effective for point-of-care information needs, particularly for discrete data like lab results and medications.
• Current infobutton utility is limited to specific electronic medical record domains and single concepts.
• Broader patient data analysis is needed to address complex health conditions.

Purpose of the Study:

• To expand the utility of infobuttons beyond discrete data points.
• To integrate a semantic web-based phenotyping engine with an infobutton framework.
• To identify and address broader patient health issues using multiple data sources.

Main Methods:

• Developed a semantic phenotyping engine leveraging the semantic web.
• Integrated the phenotyping engine with an existing infobutton framework.
• Defined and tested semantic definitions for pre-diabetes and metabolic syndrome.

Main Results:

• Successfully connected a semantic phenotyping engine with an infobutton system.
• Demonstrated the ability to identify broader patient health issues from multiple data sources.
• Validated semantic definitions for pre-diabetes and metabolic syndrome.

Conclusions:

• Connecting semantic phenotyping engines to infobuttons broadens their applicability.
• This approach can identify complex health conditions and provide relevant educational resources.
• Future work will focus on delivering timely information to patients and providers.