[Paper Review] Building a Controlled Vocabulary for Standardizing Precision Medicine Terms

Meng Wu, Yan Liu+4 more|arXiv (Cornell University)|Jul 3, 2018

Biomedical Text Mining and Ontologies1 references3 citations

TL;DR

This paper introduces the Precision Medicine Vocabulary (PMV), a controlled vocabulary built by integrating biomedical data from multiple sources using the Unified Medical Language System (UMLS). It consolidates 1,372,967 concepts and 4,567,208 terms across ten semantic types, enabling standardized data integration and normalization in precision medicine research.

ABSTRACT

Rapid advances of technology and development of research in precision medicine domain have led to the production of different types of biomedical data. Standard medical vocabularies were shown to be limited in dealing with such heterogeneous data and consequently, new controlled vocabulary for data integration and normalization has been proposed. In this study, the precision medicine vocabulary (PMV), which is a controlled vocabulary for terms used in precision medicine, is built based on the method of data integration in Unified Medical Language System (UMLS). It now covers ten top semantic types of disease, drug, gene, gene variation and so on. In total of 1,372,967 concepts and 4,567,208 terms have been integrated from widely used databases related with precision medicine.

Motivation & Objective

• To address the challenge of heterogeneous biomedical data in precision medicine by creating a standardized terminology system.
• To overcome limitations of existing medical vocabularies in handling the complexity and diversity of precision medicine data.
• To enable consistent data integration and normalization across precision medicine research by developing a comprehensive, controlled vocabulary.
• To support interoperability and semantic interoperability in precision medicine by unifying terms from multiple authoritative databases.

Proposed method

• The PMV was constructed using data integration techniques from the Unified Medical Language System (UMLS) Metathesaurus.
• Concepts and terms were aggregated from widely used precision medicine-related databases, including those for diseases, drugs, genes, and genetic variations.
• Semantic types were mapped to ten top-level categories: disease, drug, gene, gene variation, and others relevant to precision medicine.
• A normalization pipeline was applied to resolve synonymy and redundancy across source databases.
• The integration process leveraged UMLS's existing infrastructure and terminology mapping capabilities to ensure consistency and scalability.
• The resulting vocabulary was validated for coverage and coherence across key precision medicine domains.

Experimental results

Research questions

• RQ1 How can a controlled vocabulary be systematically built to standardize terms used in precision medicine across diverse data sources?
• RQ2 What is the scale and coverage of a unified vocabulary that integrates key precision medicine concepts from multiple authoritative databases?
• RQ3 To what extent does the PMV improve data integration and normalization compared to existing medical terminologies?
• RQ4 How well does the vocabulary represent core semantic types such as genes, diseases, and drugs in precision medicine?
• RQ5 Can the PMV serve as a scalable foundation for semantic interoperability in precision medicine research?

Key findings

• The PMV integrates a total of 1,372,967 unique concepts from multiple precision medicine databases.
• Over 4.5 million distinct terms were aggregated and normalized into the vocabulary, enhancing term coverage.
• The vocabulary spans ten top semantic types, including disease, drug, gene, and gene variation, ensuring broad domain coverage.
• The integration process successfully resolved term ambiguity and synonymy through UMLS-based mapping and normalization.
• The resulting vocabulary provides a scalable, standardized foundation for data integration in precision medicine applications.
• The PMV enables improved interoperability and consistency in precision medicine data exchange and analysis.