Dynamic Modeling in Medicine and Epidemiology
The Dynamic Modeling in Medicine and Epidemiology (DynaME) working group is a joint initiative of the Institute of General Practice/Family Medicine and the Institute of Medical Biometry and Statistics. Our research focuses on modeling temporal processes using longitudinal patient data, epidemiological cohort data, and complementary data from animal models.
To advance personalized medicine, we investigate methods capable of capturing individual time trajectories for each person. This enables us to identify groups with similar progression patterns. Examples include analyzing best-practice pathways within the healthcare system or uncovering underlying states in clinical studies and cohort datasets.
Methodologically, our work draws on multistate models and clustering techniques for temporal trajectories, complemented by appropriate statistical testing procedures. A central aspect of our work is to translate these methodological developments into concrete clinical applications, including those based on routine healthcare data.
- Multistate and longitudinal modelling
- Bias-aware estimation
- Integrating expert knowledge with data-driven models

Luke Breitner
- Modeling and analysis of medical data
- Machine Learning

- Modeling and analysis of healthcare pathways
- Modeling multimodal data
- Data-driven analysis of clinical routine data

- Observational studies (routine data, registries, cohort data)
- EVA4MII – Consultancy platform for the Medical Informatics Initiative

Paula Staudt
- Mathematical modeling

Parisa Westergerling
- Analysis of clinical routine data
- Medical data analysis bridging clinical practice and statistics

- CALM-QE: Privacy-compliant distributed (multimodal) analyses
- Support for clinical partners in conducting analyses
- Routine data analysis


