Building cross-thesauri with the support of UMLS

Area of Science:

• Medical Informatics
• Computational Linguistics
• Terminology Management

Background:

• Lack of a robust cross-thesaurus system hinders combinatorial terminological system development.
• Existing terminological systems require enhanced interoperability and descriptor standardization.

Purpose of the Study:

• To develop an automated tool (I-BROWSE) for generating cross-thesauri from terminological corpora.
• To improve the creation of combinatorial terminological systems and support domain experts.
• To facilitate the integration of terminological data with existing knowledge sources like UMLS.

Main Methods:

• Developed the I-BROWSE tool for analyzing terminological corpora to produce cross-thesauri.
• Integrated I-BROWSE with the Unified Medical Language System (UMLS) Knowledge Sources Server for enhanced functionality.
• Applied the tool to a corpus of 2999 surgical procedure dissections from the GALEN-IN-USE project for quality assurance.

Main Results:

• Demonstrated the utility of I-BROWSE in generating a cross-thesaurus.
• Identified significant support from UMLS for descriptors related to anatomy, pathology, and physiology.
• Showcased potential for assisting experts with less frequent or pervasive modifiers.

Conclusions:

• The I-BROWSE tool effectively aids in the creation of cross-thesauri, crucial for advanced terminological systems.
• UMLS integration provides valuable support for key medical descriptor categories.
• The tool shows promise for contributing to international terminological standards and the evolution of coding systems.