Data Science for
Digital Health
The Data Science for Digital Health (DS4DH) group focuses on developing advanced computational methods to improve healthcare. Our research lies at the intersection of Artificial Intelligence, Natural Language Processing, and Medical Informatics. We aim to extract meaningful insights from biomedical data to support clinical decision-making, enhance patient outcomes, and streamline healthcare processes.
DS4DH Research Focus
Based at Campus Biotech, University of Geneva, the DS4DH group designs novel algorithms to manage complex health data. We bridge the gap between AI methodology and actionable clinical insights through three core research streams:
Biomedical NLP
We focus on the semantic understanding of clinical text across languages. Our research spans from fundamental concept representation to biomedical language models. We apply these architectures to solve information extraction challenges, transforming unstructured records into computable healthcare information.
Deep Patient Representation
We apply deep learning to model the complexity of patient health trajectories. By using techniques like graph neural networks and temporal embeddings, we learn rich representations of longitudinal patient data to capture dynamic changes that traditional static models overlook.
Clinical Safety & Risk Assessment
We build predictive systems to improve safety in medicinal clinical research and patient care. Our research focuses on optimizing interventional clinical research through automated risk assessment, predicting adverse drug events, and modeling infectious disease dynamics in vulnerable populations.
About the Group
The DS4DH group is part of the Department of Radiology and Medical Informatics at the Faculty of Medicine, University of Geneva. Led by prof. Douglas Teodoro, our team integrates computer science and medicine to solve complex healthcare challenges.
Our Mission
Our mission is to transform raw health data into reliable clinical evidence. We specialize in three areas: structuring clinical narratives through natural language processing, modeling patient trajectories using deep representation learning, and developing predictive systems for patient safety and decision support.
Located at Campus Biotech, we work closely with the University Hospitals of Geneva (HUG) to validate our algorithms in real-world clinical settings.