Overview
Are you an aspiring data science researcher with an interest in human immunology, causal inference in dynamical systems, and/or computer vision? Would you like to apply AI and machine learning for fundamental research in biology? Then you have a part to play as a PhD candidate. By combining simulations and machine learning, you will help us develop new, innovative methods to extract knowledge from immune cell movement videos.
When thinking of cells, most people imagine static building blocks. Yet many processes in health and disease critically rely on cell movement. A key example is the defence against viruses and cancer: the cells of our immune system are strikingly motile as they navigate the maze of tissues in the human body, interact and communicate with other cells, search for signs of anomalies, and swarm to sites of infection.
Specialised microscopes allow biologists to film these processes, yielding videos that are rich in information but limited in the amount of (annotated… data for any given application. As a result, many existing AI techniques are not directly applicable to new datasets, making it very difficult to decode these videos and zoom in on the where, when, and how of key events in the data.
Our group aims to overcome these hurdles to extract meaningful insights from videos of cells. We do this by integrating data science, statistics, and AI with (bio)physics-based simulation models – using simulations to build better AI or using AI to build better simulations. In your project, you will master the arts of simulation and machine learning to help us achieve this goal.
You will be part of a larger team, for which there are currently two open positions. One PhD project will focus specifically on cell movement as a time series, developing causal inference methods for cell dynamics. The other project will be broader and, depending on your preference and developments in the field, it may focus more on the time series aspect or the computer vision aspect of video analysis. For both projects, you will first become proficient in state-of-the-art methods in the field and then improve these by integrating them with physics-based simulations.
You will spend roughly ten percent of your time (0.1 FTE) assisting with teaching activities in our department. You may, for example, be asked to provide tutorials , grade coursework, deliver presentations during classes, or supervise student projects.
We are looking for two PhD-candidates
Company:
Radboud University
Qualifications:
Language requirements:
Specific requirements:
Educational level:
Level of experience (years):
Senior (5+ years of experience)
About Radboud University Nijmegen
Radboud University is a student-oriented research university.
