Balancing Research and Data Protection: BIH Receives 1 Million Euros from the DFG for ARX-NextGen

Precision medicine tailored to individual patients and AI-driven research require large amounts of high-quality and diverse health data. Only with this data can patients’ specific characteristics be identified and taken into account in the development of new and personalized diagnostics and therapies. However, for health data to drive innovation in medicine, it must be possible to process, share, and reuse it securely without compromising patient privacy.

To this end, Fabian Prasser, a professor at the Berlin Institute of Health at Charité (BIH) and head of the “Medical Informatics” research group, is working with his team on the open-source software ARX. The software is already internationally established as a tool for anonymizing sensitive data and helps researchers assess potential privacy risks and protect information in tabular data. Now, the team plans to further develop the software as part of the “ARX-NextGen” project and has received a grant of one million euros from the German Research Foundation (DFG) to support this effort.

Data Protection for Greater Precision in Medicine

“The DFG’s funding demonstrates that technologies designed to protect patient privacy play a key role in AI-supported biomedical research and precision medicine,” says Professor Fabian Prasser. “ARX and ARX-NextGen make it possible to use sensitive data for research while simultaneously protecting the privacy of those affected. In doing so, they combine research and innovation with high European data protection standards.”

The further development of the anonymization tool acts as a catalyst for the use and reuse of data and lowers the barriers for researchers working with health data. This facilitates data analysis, particularly in the context of multicenter studies, and creates important prerequisites for the secure secondary use of data in the European Health Data Space (EHDS).

ARX-NextGen Modernizes Data Anonymization

Specifically, the research team will work on improving the integration of anonymization technologies with modern data science processes and tools. Among other things, the team plans to develop interfaces to common data formats and programming environments so that ARX-NextGen can be used directly within data science workflows. The tool will also be adapted for big data processing by expanding it to include methods for parallel and distributed data processing. Ultimately, ARX-NextGen will be made available as a web-based platform that can be integrated into medical research infrastructures and thus support, for example, the DFG’s cross-site National Research Infrastructure for Personal Health Data (NFDI4Health).

In addition to software development, the project also includes training and workshops for various user groups in data-driven medical research. The team is supported by an international advisory board comprising experts in data protection technologies, data science, and law from the United States, Canada, the Netherlands, and Germany.

The project is scheduled to last 36 months and will begin in the spring of 2027.