Semantic Interoperability and Artificial Intelligence: Carina Vorisek Takes Up New Professorship at BIH

In November 2026, Prof. Dr. Carina Vorisek took up her professorship in Semantic Interoperability and Artificial Intelligence in Healthcare and took over as head of the research group of the same name at the Berlin Institute of Health at Charité (BIH). A physician trained in Germany and the United States, she has been conducting research at the BIH since 2020 as part of Professor Sylvia Thun’s “Digital Medicine and Interoperability” research group and played a key role in establishing the national research data infrastructure NFDI4Health. We spoke with Professor Carina Vorisek about the details of her research and her goals for the new research group.
Could you describe your field of research and your areas of focus?
I develop methods for preparing and standardizing health data to make them usable for research and artificial intelligence. One area of focus is the development of data infrastructures, which we are advancing through the nationwide NFDI4Health consortium. These infrastructures help researchers find relevant data, understand its meaning, and reuse it to address new research questions.
This requires semantic interoperability: information must retain the same meaning across different institutions and IT systems. To achieve this, we develop common data models and standardized descriptions of studies and datasets—known as metadata. In doing so, we are also laying the groundwork for integration with the European Health Data Space.
Another key focus is the integration of AI. On the one hand, we are investigating how AI can support the standardization of health data. On the other hand, we are examining how data gaps and a lack of interoperability influence the validity of AI models. We pursue specific applications particularly in women’s and children’s health, for example through the MOTHER project, as well as in the context of the European Health Data Space.
Which of your past projects are particularly important to you?
The first funding phase of NFDI4Health was particularly formative for me. I helped shape the development of a national research data infrastructure and, in doing so, experienced firsthand how challenging it is to bring together different institutions, disciplines, and interests. At the beginning, there was certainly some skepticism among researchers. However, as we worked together, a shared understanding emerged of how important it is to make health data more discoverable and reusable. The prospect of the European Health Data Space further strengthened this development. It became clear to me how much a common legal framework can achieve and, at the same time, how crucial communication and interdisciplinary collaboration are. Today, I bring these experiences to the second funding phase of NFDI4Health, where I am responsible for consolidating and further developing standards and services.
Equally important to me was Midwife Roberta, a project for the digital support of pregnant women that we developed at the BIH Digital Health Accelerator. Here, I was able to collaborate with software developers from industry and gain insight into a perspective outside of academic research. The focus was very specifically on what users need and how an idea can be turned into an application that can be used in everyday healthcare practice. That experience helps me a great deal today in designing research projects from the very beginning with an eye toward their future implementation.
What are your plans for the research group?
With my research group, I want to play an active role in shaping Germany’s data infrastructures for the European Health Data Space. In doing so, we aim to develop and test methods that will allow health data to be made usable and standardized more efficiently, including with the help of AI. It is crucial that the data remain reliable and comply with the FAIR principles, meaning they must be findable, accessible, interoperable, and reusable.
Building on this, we aim to specifically identify and close data gaps, particularly in women’s health and rare diseases. We’re also interested in determining which other diversity characteristics have not been adequately captured to date and what this means for research and AI applications.
Another goal is to develop practical tools based on our research. To this end, I would like to expand our collaboration with industry and startups and strengthen international partnerships, including through my connections in the U.S. Together, we want to explore how data infrastructures can support the development and independent validation of AI applications across national borders.
As a physician, it is particularly important to me to further strengthen the bridge to the clinical setting: Clinical questions should shape our research, and our findings should be applied in clinical practice. For me, this also includes training physicians in medical informatics and opening up opportunities for them to collaborate with computer science and data science.
How does your research align with the BIH’s translational mission? How might patients benefit from your work in the future?
Bringing research into clinical care is my greatest motivation. I have worked in both outpatient and inpatient settings in Germany and the United States. These experiences help me understand different care processes and, when developing new solutions, to adopt the perspective of the people who will eventually use them.
This also shapes our projects. In NFDI4Health, we are developing a research data infrastructure with the goal of making its services permanently available. That’s why we incorporate the needs of users from the very beginning: data-holding organizations, researchers, and members of the general public, who can learn about health studies through the Health Study Hub. The benefit for patients here stems primarily from the fact that existing data can be better utilized to address new research questions.
MOTHER shows how we use this infrastructure: We prepare longitudinal data accessible through NFDI4Health according to the FAIR principles and apply AI methods. We investigate how maternal and early childhood factors relate to obesity and metabolic changes in girls and boys, what differences exist between the sexes, and how the underlying data affect the results. The aim is to close sex-related gaps in data and knowledge, enabling earlier and more targeted prevention.
For me, translation therefore begins with the research question: What are the practical needs, who do we need to involve, and how can a solution be used in the long term? In this way, our work ties directly into the BIH’s translational mission. Our goal is to use data infrastructures and methodological research to create reliable tools that can support clinical decisions and improve patient care.