SYNTHIA will be presenting at both TDPD 2026 and the OpenDP Differential Privacy for Health and Genomics Workshop, 1-2 June 2026, in Boston, USA.
SYNTHIA partner Bogdan Kulynych, Lausanne University Hospital (CHUV), will present in the session "Reconciling Differentially Private Medical Data and Model Sharing with Data Protection".
Bogdan will present the framework developed through publication "Unifying Re-Identification, Attribute Inference, and Data Reconstruction Risks in Differential Privacy", published in NeurIPS 2025.
The publication highlightes a framework that tackles a key challenge in privacy-preserving healthcare AI: ensuring that synthetic data and machine learning models can be shared with provably low risks of re-identification, attribute inference, and data reconstruction, while remaining useful for research. By introducing a more precise and unified way to quantify these privacy risks under differential privacy, the work helps bridge the gap between theoretical privacy guarantees and the practical risk measures referenced in data protection and regulatory guidance. This represents an important step towards making privacy-preserving synthetic data both scientifically rigorous and suitable for real-world healthcare applications.
About the conferences
The OpenDP Differential Privacy for Health and Genomics Workshop brings together researchers, practitioners, and policymakers working to advance the responsible use of differential privacy in health and genomics. The workshop is part of the broader OpenDP initiative, a community-driven effort to develop trustworthy open-source tools that enable privacy-preserving analysis of sensitive personal data. By providing accessible implementations of differential privacy algorithms, OpenDP supports government, industry, and academic institutions in sharing and analysing sensitive data while protecting individual privacy. The initiative continues to grow through collaboration across academia, industry, and the public sector, helping translate the latest advances in differential privacy into practical tools for real-world applications.
TDPD 2026 (Theory and Practice of Differential Privacy) brings together researchers and practitioners to discuss the latest theoretical advances and practical applications of differential privacy. TDPD 2026 is co-located with the OpenDP Differential Privacy for Health and Genomics Workshop and the Foundations of Responsible Computing Conference, creating a unique opportunity for interdisciplinary collaboration on trustworthy, privacy-preserving computing and responsible AI.
