JMIR Publications

ABSTRACT

Background:

Efficiently finding and exploring relevant health studies is critical for informed evidence-based healthcare. However, study information remains distributed across multiple resources, hindering interoperability, search, and reuse. Enhancing the findability of study data is a key challenge in promoting the FAIR principles in health research.

Objective:

This work aimed to improve the findability and comparability of health studies by developing a semantically enriched graph-based framework that supports intuitive search and exploration for diverse stakeholders, including clinicians, researchers, and patients.

Methods:

We developed the BRAinS-Graph (“Biomedical Knowledge Graph for Recommending and Analysing Health Studies”), a semantically enriched knowledge base that integrates data from ClinicalTrials.gov, the Portal for Medical Data Models, the Unified Medical Language System (UMLS), and Medical Subject Headings (MeSH) into a single graph database. The framework applies an extract–transform–load (ETL) process to integrate heterogeneous data structures and link related information across study resources and biomedical ontologies.

Results:

The BRAinS-Graph supports fine-grained, semantic searches across study metadata, eligibility criteria, and structural properties. Use cases illustrate its potential for clinicians, patients, and researchers, including analyses of study type distributions for meta-analyses and the identification of studies relevant for individual patients.

Conclusions:

By integrating heterogeneous study data into one interconnected knowledge base, the BRAinS-Graph improves the findability, accessibility, and reusability of study information, thereby advancing the FAIRness of health research. The present work establishes a foundation for graph-based study recommendation systems and cross-institutional research infrastructures.