Netcompany has published the SYNTHEMA SMPC federated learning app on the Flower Hub. It enables secure aggregation of model updates while keeping data private.
👉 Explore the app: flower.ai/apps/synthem...
Research on rare hematological diseases depends on people committed to complex challenges. Within SYNTHEMA, women contribute across clinical research, data science, technology and project work. Today, on March 8, we spotlight their contributions and give them the recognition they deserve.
At #SYNTHEMA we use AI to fill data gaps & improve care for rare blood disorders. Focusing on Sickle Cell Disease & Acute Myeloid Leukaemia, we integrate clinical, genetic & imaging data for improved patient outcomes.
👉 synthema.eu
#RareDiseases #Haematology
28/02 is Rare Disease Day. It is about connection across countries, disciplines and communities. Progress in rare diseases depends on visibility, collaboration and responsible data use.
At SYNTHEMA, we support cross-border research in rare hematological diseases synthema.eu
Interested in synthetic data and AI-powered breakthroughs in hematology? Visit SYNTHEMA’s publications section to explore the latest advancements by our team in data privacy, federated learning, AI for rare blood diseases, and more.
👉 Start reading: synthema.eu/publications/
#SYNTHEMA
SYNTHEMA uses smart AI to help scientists study diseases like leukaemia and sickle cell without using real patient data.
These models learn from medical records and create new, realistic examples, such as lab results or scans, while keeping privacy fully protected.
👉 synthema.eu
Scientific progress depends on who gets to take part. In #SYNTHEMA, women contribute as researchers, clinicians and data specialists across all stages of the work. This day draws attention to how scientific careers are accessed: synthema.eu/2025/02/11/b...
A worth conference to keep an eye on, with a focus on AI applications and concrete clinical cases in rare diseases, alongside patient perspectives.
At SYNTHEMA, we support cancer research in rare haematological diseases such as Acute Myeloid Leukaemia by developing privacy preserving synthetic clinical and genomic data, helping survival and treatment response research where patient numbers are limited.
👉 Learn more: synthema.eu/use-cases/
It’s not enough for synthetic data to look real, it must be clinically useful. #SYNTHEMA tests if it can predict outcomes, reveal genetic risks & support real diagnoses.
Synthetic data means real impact.
👉 synthema.eu
#SyntheticData #AIHealthcare
Rare disease research can’t be equitable if data stay scarce or siloed. At SYNTHEMA, we tackle these barriers with GDPR-compliant AI and synthetic data, expanding the evidence base for rare hematological diseases so innovation can benefit patients everywhere.
👉 synthema.eu
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:
📩 by email: f981d420.sibforms.com/serve/MUIEAH...
💼 on LinkedIn: www.linkedin.com/newsletters/...
These digital features significantly improved overall survival predictions, raising the concordance index from 0.68 to 0.87. Such explainable AI tools provide the precision needed for personalised haematology.
Read the full research at synthema.eu/publications/
The Synthema and GenoMed4All consortia have developed an AI framework to refine myeloid neoplasm diagnosis and prognosis. By extracting features from bone marrow images, the team achieved diagnostic accuracy exceeding 90%.
#SYNTHEMA is part of the #HealthData4EU cluster, an alliance of EU-funded projects addressing fragmentation in health data across Europe through synthetic data generation, federated learning, and privacy-preserving technologies.
Laarn more: synthema.eu/healthdata4eu/
The platform is currently for research purposes only, designed to help clinicians explore the Digital Twin concept in practice.
📑Access the full poster abstract from the 66th ASH Annual Meeting here: synthema.eu/publications/
By simulating a patient’s natural history, GEMINI provides evidence-based estimates for:
✅ Survival probability
✅ Leukemic evolution risk
✅ Optimal timing for interventions like transplantation
How can AI improve precision medicine in hematology?
The GEMINI Digital Twin platform, developed by the #SYNTHEMA & #GenoMed4Al consortia, integrates multimodal real-world data from 22,080 MDS patients to personalize clinical decision-making.
#MDS #Hematology #DigitalTwin
We start this year sharing that registrations are now open for ECRD 2026, taking place on 3–4 June 2026 in Prague and online. The conference addresses European policy priorities on rare diseases, including rare hematological diseases.
Early-bird registration is available until 26 February 2026.
A short holiday break is approaching. Wishing partners, colleagues and everyone following #SYNTHEMA a peaceful holiday period, whatever it may look like for you.
Thank you for being part of a community committed to responsible research in rare hematological diseases.
synthema.eu
PW FedAvg, a new method by UPM researchers, improves treatment effect estimation in federated healthcare settings with imbalanced treatment distributions. It reduces causal inference errors where traditional FL methods fall short.
👉 synthema.eu/publications/
#SYNTHEMA #FederatedLearning
#SYNTHEMA joined the i~HD Annual Conference, with Elisabetta Mezzalira (University of Padova) representing the project. The event featured the presentation of the #HealthData4EU white paper “Building Trust in Synthetic Health Data”.
#SyntheticHealthData
How similar is synthetic tabular data to the real thing?
This new approach uses KL and JS divergences to evaluate full joint distributions — not just feature by feature. A more reliable way to validate synthetic datasets in practice.
👉 synthema.eu/publications/
#SyntheticData #DataValidation #AI
Over 450 haematological diseases affect blood cells, the lymphatic system & coagulation, posing major challenges to public #HealthSystems.
#SYNTHEMA is developing a secure platform for synthetic #ClinicalData to enable GDPR-compliant research in rare #HaematologicalDiseases.
👉 synthema.eu
Can synthetic tabular data be more accurate?
This new model by researchers at Universidad Politécnica de Madrid combines a Variational Autoencoder with a Gaussian Mixture Model. Results outperform #CTGAN and #TVAE, especially on medical datasets.
👉 synthema.eu/publications/
#SYNTHEMA
The paper “Distributing trust: A P2P SMPC protocol for secure federated learning” by Petros Demetrakopoulos, Themistoklis Anagnostopoulos and Sofia Tsekeridou (Netcompany RID, #SYNTHEMA partner) will be presented at #FLLM2025 on 27 Nov at 15:30.
synthema.eu
#HealthData4EU Cluster is at #EBDVF25 this week. SYNTHEMA isn’t on site, but we’re glad to support tomorrow’s session in Copenhagen. If you're there on 14 Nov, join “Ecosystems of Innovation in Healthcare” from 09:00–10:30, Room 2.B — a chance to hear how our partners advance ethical health data use
How well does synthetic data generation work when spread across multiple hospitals? This study on #AML shows that federated learning protects privacy, though fidelity takes a hit. A key step in real-world AI for rare diseases.
👉 synthema.eu/publications
@vicomtech.bsky.social
#SYNTHEMA
Two productive days in San Sebastian for the #SYNTHEMA consortium have come to an end. Lively discussions, clear goals for the final year: a recap publication on synthetic data in haematology, joint training with @erneurobloodnet.bsky.social, new papers, and the SYNTHEMA platform launch
synthema.eu
The SYNTHEMA consortium is meeting in San Sebastian, Spain, to mark the project’s third anniversary and kick off discussions for the final year ahead. Together, we’re committed to delivering results in secure, AI-powered health data for rare hematological diseases.
👉 synthema.eu
