A framework for distributed mediation of temporal-abstraction queries to clinical databases

Area of Science:

• Medical Informatics
• Distributed Systems
• Clinical Data Management

Background:

• Medical applications require complex temporal queries on clinical data and knowledge.
• Integrating heterogeneous data sources and medical knowledge is challenging.

Purpose of the Study:

• To develop a distributed system integrating medical knowledge bases with time-oriented clinical databases.
• To enable answering complex temporal queries on raw clinical data and its abstractions.

Main Methods:

• Designed a virtual medical-record interface for heterogeneous clinical data sources.
• Developed the ALMA service for computing complex time-oriented medical queries.
• Implemented the IDAN mediator to integrate clinical data and medical knowledge.
• Created an expressive query language and evaluated the IDAN/KNAVE-II system with clinicians.

Main Results:

• The full IDAN architecture was implemented and integrated with the KNAVE-II visual interface.
• The IDAN/KNAVE-II system significantly enhanced the speed and accuracy of complex query answering.
• Clinicians using the system showed marked improvement over paper or spreadsheet methods.

Conclusions:

• The IDAN architecture demonstrates the feasibility of distributed integration of diverse medical data and knowledge.
• The modular IDAN system shows potential for automating clinical tasks like diagnosis, monitoring, and quality assessment.