Young Researchers

Rosni Vasu, PhD

Dr. Rosni Vasu is a Postdoctoral Researcher in the Department of Clinical Research at the University of Bern. She completed her PhD in Informatics at the University of Zurich, where she developed AI methods for scientific discovery, particularly approaches that enable researchers to generate and evaluate new hypotheses and navigate complex bodies of evidence.

Building on this foundation, her current work advances AI for health and societal good. She develops AI systems that bring together diverse scientific and clinical evidence to support decision-making in healthcare and drug development, helping to identify promising research directions and enable more efficient, evidence-driven discovery.

Rosni Vasu emphasizes the importance of developing approaches that are not only technically robust, but also aligned with clinical and societal needs. By focusing on applications with tangible and visible impact, she contributes to ongoing efforts to translate advances in AI into meaningful benefits for research, healthcare, and society.

Looking ahead, she aims to focus her research on real-world needs in healthcare, developing practical AI-driven solutions that can be applied in research and practice.

Healthcare is far more complex than any single technology, and the role of AI is to support, not replace, how we understand and use evidence

“

Vera Bernhard

Vera Bernhard is a Data Scientist at STRIDE-Lab, the Medical Data Science group at the Department of Clinical Research. Her interdisciplinary interests led her to pursue a Master’s degree in Computational Linguistics & Language Technology and Informatics.

During her studies, she focused on applying and fine-tuning (large) language models and general machine learning methods for a diverse set of tasks, including speech recognition, medical report generation, and text classification. At the same time, she discovered her passion for making large data available and discoverable by building web apps from scratch and using human-computer interaction methods to make them intuitive and user-centred.

After initially joining STRIDE-Lab as a student assistant and transitioning into a data science role after graduation, Vera combines her expertise in machine learning and web development to work with biomedical text data. Currently, she develops pipelines to automate systematic reviews in the domain of psychedelic therapy for psychiatric diseases. Therefore, she applies language models to classify and extract relevant information from clinical study publications, making the insights accessible through an interactive web interface.

Looking ahead, Vera aims to continue translating clinical needs into technical solutions, bringing together perspectives from different fields of research and making large-scale biomedical data more accessible. She is especially looking forward to moving from meta-science to working directly with clinical data.

“Bring medical and technical voices together early: without medical perspective, you might solve the wrong problem; without technical perspective, you might define one you can’t solve.”