The SYNTHEMA & ERN-EuroBloodNet Educational Program on Artificial Intelligence and its applications in hematology has been conceived by the Synthema Educational Scientific/Technical Committee and targets both the public-at-large and expert audiences through a dedicated webinar series. The programme aims to increase awareness and understanding of cutting-edge AI approaches in hematology, with a particular focus on synthetic data generation and federated learning methodologies, demonstrated through two clinical use cases: Acute Myeloid Leukemia (AML) and Sickle Cell Anaemia (SCA).

Program

The SYNTHEMA & ERN-EuroBloodNet Educational Program on Artificial Intelligence in hematology for the public-at-large provides an overview of SYNTHEMA's mission and its key applications of Artificial Intelligence in hematological research and care.

The program aims to:

• Provide a shared understanding of the common infrastructure for synthetic data generation within ERN-EuroBloodNet. ​
• Clarify ethical and regulatory frameworks, including Medical Device Regulation (MDR) and the EU Artificial Intelligence Act. ​
• Address technical foundations such as data standardisation, interoperability, anonymisation, and synthetic data methodologies. ​
• Demonstrate real-world applicability through concrete disease-specific use cases. ​
• Strategically position ERN-EuroBloodNet as a coordinated, federated research network aligned with European regulatory and data standards. 

It introduces core concepts underpinning SYNTHEMA's approach, including the role of AI in rare hematological diseases, synthetic data generation and federated learning as solutions to data scarcity and fragmentation, and the importance of data standardisation and cross-border interoperability. These topics are illustrated through two clinical use cases-Acute Myeloid Leukemia (AML) and Sickle Cell Anaemia (SCA)-and showcase how AI-driven, privacy-preserving and multi-modal approaches can support disease understanding, diagnosis, and clinical research. Mentioned topics are divided into those 4 different sessions, as follows:

• Session 1: SYNTHEMA - Infrastructure for SDG in ERN-EuroBloodNet

Date: 8 May from 12:30 to 14:00

Speakers: Federico Alvarez & María del Mar Mañú

Introductory session presenting the federated ERN-EuroBloodNet infrastructure for synthetic data generation, with a focus on strategic positioning, governance, and technical architecture.

• Session 2: Regulatory on Medical Devices, AI Act, Privacy-Preserving and Ethical Considerations

Date: 15 May from 12:30 to 14:00

A session structured in two blocks: ​

1.Overview of ethical and regulatory considerations, including MDR and the EU AI Act, and their implications for software, platforms, and data-driven tools in healthcare.​

Duration: 45 minutes​

Speaker: Nathan Lea​

2.Privacy Preserving in Federating Learning​

Duration: 45 minutes​

Speaker: Rudolph Mayer & Sofia Tsekeridou

• Session 3: Data Quality, Standardisation, Synthetic Data Generation, anonymization and interoperability

Date: 22 May from 12:30 to 14:45

A technical session structured in three blocks: ​

1.Data quality, data standardisation and interoperability (OMOP, data models)​

Duration: 45 minutes​

Speakers: Sara Reidel, Giulio Spinozzi​

2.Synthetic data generation and anonymisation methodologies​

Duration: 45 minutes​

Speakers: Sofia Tsekeridou, Imanol Isasa.

3.Building Trust in Synthetic Health Data: Methodological and Acceptance Challenges across Seven European Projects​

Duration: 45 minutes​

Speakers: Inês Sousa, Jan Ramon, Rafael Redondo, Max Salmi. 

• Session 4: Opportunities of Synthetic Data Generation in Rare Haematological Diseases

Date: 29 May from 12:30 to 14:30

Applied session showcasing disease-specific use cases in Sickle Cell Disease (SCD) and Acute Myeloid Leukaemia (AML)​

1.SCD​

Duration: 60 minutes​

Speakers: María del Mar Mañú, Raffaella Colombatti​

2.AML​

Duration: 60 minutes​

Speakers: Eleonora Iascone​