Overview
SYNTHEMA is a cross-border hub that develops and validates AI techniques for anonymisation and synthetic data generation in rare haematological diseases. The company aims to generate reliable, high-quality synthetic data to create virtual patients for diagnostic enhancement, treatment assessment, and outcome prediction in haematology, addressing data silos scattered across European clinical sites and registries.
In the news
- Synthetic data is becoming an increasingly important part of the conversation around the future of digital health, from improving access to data and supporting AI development to enabling more privacy-preserving research. Last year, experts from the BDVA - Big Data Value Association published the white paper “Synthetic Data in Healthcare.” The paper examines how synthetic data can support healthcare innovation while addressing persistent challenges related to regulation, governance, representativeness, benchmarking and technical
- Sickle cell disease presents every challenge in rare diseases.” In the next edition of SYNTHEMA | The Final Stretch, María del Mar Mañú Pereira from Vall d’Hebron Institute of Research (VHIR) - Vall d'Hebron Institut de Recerca explains why Sickle Cell Disease was selected as one of the project’s clinical use cases. Watch the interview to hear more: https://lnkd.in/eFZaJVyk More about SYNTHEMA: https://synthema.eu/
- Proud to see SYNTHEMA represented in this work through our partners Petros Demetrakopoulos and Sofia Tsekeridou from Netcompany, who contributed as co-authors. SYNTHEMA is also part of the Flower Hub ecosystem with an application developed within the project. The SMPC-FL app implements a peer-to-peer Secure Multi-Party Computation protocol for federated learning, allowing model updates to be aggregated without exposing individual participants’ contributions. Explore the SYNTHEMA application on Flower Hub:
- SYNTHEMA is entering its final stretch. Over the coming weeks, we will be sharing contributions from our partners, looking at the results of their work and the progress made across the project. We begin with Rudolf Mayer and Daniela Martínez Duarte from SBA Research, who explain how privacy-by-design, federated learning, SMPC and differential privacy help protect patient information throughout synthetic data generation. Swipe through to see how we’re addressing one of synthetic data’s most important challenges. More about SYNTHEMA:
- Save the date: 3 November 2026 Our partner Università degli Studi di Padova (UNIPD) / University of Padua is organising a dedicated SYNTHEMA Symposium in Padua, focused on real-world data, federated learning, synthetic data and the future of data-driven medicine. More details soon. https://synthema.eu/
- If our four-part training series on synthetic data in rare haematology was not on your radar earlier this year, the full programme remains open-access. Across four sessions, leading clinical and technical experts share practical methods for training generative models locally, keeping sensitive patient records fully secure. The complete series covers: ‣ How federated infrastructure trains models directly across local hospital nodes ‣ Navigating high-risk AI Act classifications and GDPR compliance ‣ Standardising complex health
- Patient involvement is an essential part of rare disease research and care. For SYNTHEMA, working within the rare haematological disease ecosystem and alongside ERN-EuroBloodNet (European Reference Network on Rare Hematological Diseases) also means communicating clearly how synthetic data can support research while protecting patient privacy, and ensuring that patients and patient organisations understand how these approaches can be used in research. Here is a useful overview from JARDIN Joint Action on how patients and patient
- Subscribe to the SYNTHEMA newsletter to receive key project updates and announcements about important activities, all in one place. Clear, relevant, and focused on what matters. Choose the format that suits you best: 📩 by email: https://lnkd.in/epezaNg4 💼 on LinkedIn: https://lnkd.in/eFZESGdB
Something wrong or missing? Send an update. Fixed within 24 hours.