Faculty of Medicine

Synthetic Data Day

AI-generated Illustration of Synthetic Medical Data

Synthetic patient data – these are AI-generated datasets that mimic the underlying real patient data so closely that the statistical properties of the original data are retained, but no personal information is available. Patient data "synthesized" in this way has the potential to solve a pressing problem in medical research. Namely, to make patient data widely available for research while fully and irreversibly protecting patient privacy.

What seemed like an insurmountable dilemma just a few years ago is now becoming a reality thanks to AI models. The Faculty of Medicine, the University of Bern and the Inselspital want to jointly explore the topic of "synthetic data" and lay the necessary foundations. The Synthetic Data Day on May 27, 2024, was the first step in this direction.

First symposium on synthetic data

The presentations by national and international experts provided exciting insights into the potential of synthetic data, its current areas of application, and existing research needs. All presentations are available as video recordings.

Synthetization of Complex Data and the Swiss Data Anonymization Competence Center

Prof. Dr. Matthias Templ

Prof. Dr. Matthias Templ, University of Applied Sciences and Arts Northwestern Switzerland & Swiss Data Anonymization Competence Center

Expanding AI Capabilities through Synthetic Data

Prof. Dr. Randi Foraker

Prof. Dr. Randi Foraker, Director of the Center for Population Health Informatics (CPHI), Washington University, St. Louis, USA

Synthetic Data in clinical research: are we there yet?

Ass. Prof. Dr. Jean Louis Raisaro

Ass. Prof. Dr. Jean Louis Raisaro, Head of the Clinical Data Science Group, CHUV-UNIL Biomedical Data Science Center

Synthetic Data: Powerful Creation, Not Second Rate Copy

Prof. Dr. Mihaela van der Schaar

Prof. Dr. Mihaela van der Schaar, John Humphrey Plummer Professor of Machine Learning, AI, and Medicine and Director of the Cambridge Center for AI in Medicine, UK

Comparing LLMs and GANs for Tabular Data Generation

Prof. Dr. Murat Sariyar

Prof. Dr. Murat Sariyar, Lecturer (University of Bern) and Head of Data Science Specializations in Medical Informatics (Bern University of Applied Sciences)

Synthetic Health Data: facilitating sharing and enhancement

Lucy Mosquera

Lucy Mosquera, Senior Director of Generate Operations and Data Science, Aetion (formerly Replica Analytics), Ottawa, Canada