STO: Stroke Ontology for Accelerating Translational Stroke Research

Area of Science:

• Biomedical Informatics
• Neuroscience
• Cardiovascular Diseases

Background:

• Ontology-based annotation accelerates disease knowledge analysis.
• Existing ontologies lack comprehensive representation for the stroke domain.
• There is a need for structured knowledge representation in cardiovascular diseases, specifically for stroke.

Purpose of the Study:

• To develop the first multiple-view ontology for the stroke domain.
• To create a tool for formalizing and standardizing heterogeneous stroke-related data.
• To incorporate diverse user perspectives (neuroscientists, molecular biologists, clinicians) into a unified knowledge base.

Main Methods:

• The stroke ontology (STO) was developed following the ontology development life cycle.
• Protégé ontology editor and Ontology Web Language (OWL) were utilized for construction.
• Evaluation included structural/functional features, expert review, and competency questions.

Main Results:

• STO encompasses major biomedical and risk factor concepts related to stroke.
• Concepts are enriched with synonyms, definitions, and references for clarity.
• Natural language processing evaluation yielded high precision (0.94), recall (0.80), and F-score (0.78).
• Competency questions demonstrated STO's ability to answer expert queries based on evidence.

Conclusions:

• STO is the first multiple-view ontology for brain stroke, enabling data representation, formalization, and standardization.
• The ontology serves as a valuable tool for the stroke research community.
• Community input is encouraged to enhance the current draft version of STO.