I am a mathematician with an interest in how data from routine healthcare can be used to answer questions about disease and medicine use. My research spans a broad range of epidemiological areas, such as infectious diseases, cancer and diabetes, as well as pharmacoepidemiology. I am particularly curious about how larg language models can unlock the information held in free-text medical reports. I am currently a biostatistician in the Department of Clinical Epidemiology at Aarhus University.
My own research focuses on natural language processing (NLP) methods for structuring information from free-text medical reports, and on how these methods can improve phenotyping and outcome ascertainment in epidemiological research.
Responsible for choosing sufficient statistical methods, defining common data
models, performing data management and the analysis itself including artifacts like
publication-ready tables and advanced visualizations. Defining cross-country consistent data modelling and processing in relation to European projects, involving multiple data partners.
