Europe's Largest Research Infrastructure for Curated Medical Data Models with Semantic Annotations

Area of Science:

• Medical Informatics
• Health Data Management
• Bioinformatics

Background:

• Structural metadata from clinical studies and healthcare systems is largely inaccessible to researchers.
• A significant gap exists in the availability of standardized and annotated medical data models.

Purpose of the Study:

• To present an overview of the content within the Portal of Medical Data Models (MDM Portal).
• To highlight the MDM Portal's role as a European information infrastructure for research and healthcare.

Main Methods:

• The MDM Portal curates and semantically annotates medical data models using expert knowledge.
• Users can search, view, discuss, and download these medical data models.

Main Results:

• The portal contains 25,257 data models with over 1.6 million Unified Medical Language System (UMLS) codes assigned.
• "Clinical trial" and "breast neoplasms" are the most frequent keywords.
• Semantic annotations cover over 800,000 elements, with data primarily in English and German.

Conclusions:

• The MDM Portal is Europe's largest collection of semantically annotated medical data models.
• It facilitates increased compatibility of medical datasets.
• It serves as a substantial expert-annotated corpus for natural language processing applications.