Building on the OMOP 101 session, this presentation shifts from why OMOP to HOW to implement it. Organizations looking to adopt the OMOP Common Data Model (CDM) need to understand the technical, organizational, and collaborative steps involved.

Using Helsinki University Hospital (HUS) as a case study, this session provides a presentation of HUS roadmap to OMOP adoption—from data ingestion and ETL to quality control, the first federated real-world study to predictive models using advanced swarm learning.

From Theory to Practice:

• Preparation & Planning – Assess existing data infrastructure, define objectives, and engage stakeholders.
• Data harmonization in three steps:
  • Data discovery – Prepare raw data (Data analysts)
  • Concept mapping of data – identify OMOP concepts for the source data (Clinical experts)
  • ETL (Extract, Transform, Load) – Harmonize data into the OMOP CDM format (Data engineers)
• Data Quality & Validation – Implement verification and iterative testing to ensure reliability.
• Establishing analytical capabilities
  • Infrastructure & Tools – Leverage OHDSI’s open-source ecosystem, including Atlas and WhiteRabbit.
  • Training & Community Engagement – Build a multidisciplinary team and connect with OHDSI working groups and local country node.
  • Pilot Testing & Evaluation – Conduct initial studies, refine processes, and scale implementations.
  • Advanced OMOP: AI & Swarm Learning.

Helsinki University Hospital’s OMOP Journey:

HUS has successfully transformed its healthcare data into OMOP CDM in close collaboration with FinOMOP and the other university hospitals in Finland, enabling:

• Efficient data integration through version controlled national and international standards.
• Participation in federated studies to generate real-world evidence; EHDEN, DARWIN, OHDSI, …
• Advancing AI-driven research through swarm-learning – a form of federated machineleraning –   to collaboratively build decentralized predictive models.

The Value of the OHDSI Community. Joining OHDSI means access to:

• A global research network for large-scale studies.
• Open-source tools & best practices for OMOP adoption.
• Education, training, and mentorship to accelerate both research and implementation.