Josefine Andreasson is a PhD student in epidemiology and population studies (EPI@Lund), and her research is about why some people are more susceptible to respiratory infections than others. She focus on identifying underlying causes, analyzing consequences and evaluating preventive measures.
Why did you apply for the course?
- I have been fascinated by the possibilities of machine learning and I am inspired by colleagues who use these methods in their research.
Machine learning is a field that is developing at a furious pace and has enormous potential in epidemiology. When the course “Machine Learning in Epidemiology” was advertised by EpiHealth, I saw a perfect chance to get a practical introduction, learn the basics and understand the language – both to be able to apply the methods in my own research and to facilitate collaborations with experts in the field.
What was the most important lesson?
- That machine learning is more accessible than I thought! The course provided a broad introduction to different methods – everything from classic regression models to advanced neural networks. We also got to try applying them in practice, which will really help when I implement the methods on my own. In addition to new knowledge, I take with me the insight that these tools are fully possible to use even for researchers without a technical background and that they can open completely new doors for epidemiological research. The idea workshop was in full swing on the train home, and I am now looking forward to implementing what I learned in my own research.
What are your take aways from the meeting on the last day?
- That the Swedish cohort landscape is an enormous resource that enables significant research in many different areas. The meeting also provided a fantastic opportunity to meet other members of the EpiHealth network and exchange ideas.
I left with both new knowledge and lots of inspiration.
