Creating a global single source of truth' for medications data using AI, blockchain, and pharmacogenomics
Coach Name
Jordi Bosch Garcia
EU Organization
Dynaccurate SARL
Members
- Dermot Doyle
- Silvio Cardoso
CA Organization
Pillcheck
Members
- Veronika Litinski
- Ruslan Dorfman
Project Overview
MDIP addresses a critical gap in global healthcare: the lack of interoperable, accessible medications data across countries, which directly impacts patient safety and public health outcomes.
The project created a cross-border, open-access medications database linking authoritative datasets from the EU, UK, US, and Canada with pharmacogenomic (PGx) risk flags—highlighting drugs that may cause adverse reactions based on a patient’s genetic profile.
By combining AI-driven semantic interoperability, ontology mapping, and blockchain-based provenance, MDIP established a “single source of truth” for medications data. The system demonstrates how public data can be enriched and reused to improve safety, reduce costs, and enable international data collaboration.
Methods and approaches
AI-Driven Semantic Interoperability & Data Harmonisation
MDIP developed a methodology to:
- Collect and ingest medications datasets from public authorities across multiple jurisdictions
- Map equivalent medications using ATC codes and active substances
- Enrich datasets with pharmacogenomic flags provided by Pillcheck
- Use ontology-based modelling to create a unified, interoperable dataset
This approach enabled the creation of a dynamic, cross-border medications knowledge base despite differences in national data formats and languages.
Blockchain-Based Provenance & Open Data Platform
The project implemented a blockchain layer to ensure:
- Data traceability and integrity
- Transparent provenance of sources
- Trust in cross-border data exchange
All processed data was published via an open-access web platform (www.mdip.eu) and shared through GitHub, demonstrating how public sector data can be reused at scale.
Key Achievements
Creation of a live, searchable international medications database (www.mdip.eu).
Integration of datasets from EU, UK, US, and Canada regulators.
Collection of 950,197 medication records, with 930,871 valid for mapping.
Identification of 223,152 medications (≈24%) with pharmacogenomic risks.
Successful demonstration of AI + ontology + blockchain architecture.
Publication of datasets and code via open-source GitHub repository.
Delivery of full interoperability workflows for cross-border medication matching.
Execution of 46 stakeholder interviews across healthcare, pharma, and policy sectors.
Dissemination through blogs, videos, conferences, and stakeholder roadshows.
Impact & Results
Scientific Impact
MDIP demonstrates the real-world feasibility of achieving semantic interoperability across highly heterogeneous healthcare datasets, a long-standing challenge in the field. By aligning multiple national data sources into a unified structure, the project lays the groundwork for the development of global medication knowledge graphs. In doing so, it also reinforces the importance of pharmacogenomics in advancing personalized medicine and improving patient safety through more informed, data-driven insights.
Societal Impact
The project contributes to greater awareness of potentially life-threatening adverse drug reactions linked to genetic factors, an area often overlooked in standard prescribing practices. By making this information accessible, MDIP empowers both clinicians and patients to make better-informed treatment decisions. At a broader level, it supports public health systems in reducing medication-related risks and associated costs, ultimately contributing to safer and more efficient healthcare delivery.
Economic Impact
MDIP highlights the value of interoperable healthcare data as a driver for innovation across the life sciences sector. By enabling more seamless data integration, the project opens new opportunities for data-driven services, including clinical research, pharmaceutical development, and healthcare analytics. At the same time, it sheds light on inefficiencies in national medication procurement and data management systems, pointing to areas where improved interoperability can lead to significant cost savings and operational improvements.
Economic Impact
The collaboration between Dynaccurate and Pillcheck represents a strong synergy between AI engineering and clinical expertise, demonstrating the added value of transatlantic partnerships. This cooperation has already extended beyond the project through joint funding applications and participation in public tenders, and it establishes a scalable model for cross-border health data interoperability. As such, MDIP contributes to strengthening long-term EU–Canada collaboration in digital health innovation.
Publications and Open-Source Contributions
- Open-access MDIP platform: www.mdip.eu
- GitHub repository with datasets and code
- Blog articles, videos, and educational content on pharmacogenomics and interoperability
- Conference presentations (Luxembourg Digital Health Conference, ENDORSE 2025)
- Data contributions to EU-level initiatives (eHDSI, SNOMED, ELIXIR Europe)
Future directions