We offer 3 fully funded PhD positions to work on research methodology with experts in a great medical research environment. Interested in one of these projects? Send us your motivational letter and up-to-date CV via the website of the UMC Utrecht: more info in links below.
Deadline for applicants: May 15th 2025.
———
PhD position 1: Methods for foreseeing temporal performance deterioration of AI-based prediction models
It is well known that prediction models tend to lose some of their predictive performance over time. This project explores the frequency at which AI-based prediction models need testing, and to what degree deteriorating predictions can be anticipated and prevented.
Supervisory team: Anne de Hond, Nicole Erler, Carl Moons, Maarten van Smeden
More info & apply at: PhD position Temporal Performance Deterioration in AI-based Prediction Models ———
PhD position 2: Causal inference methods to counter the prediction paradox in deployed AI-based prediction models
The prediction paradox describes the phenomenon that predictions can influence behaviors that in turn invalidate predictions. In this project the candidate will help us quantify the effects of the prediction paradox and develop new methods to counter its potential detrimental effects when updating deployed AI-based prediction models.
Supervisory team: Wouter van Amsterdam, Oisín Ryan, Ewout Steyerberg, Maarten van Smeden
More info & apply at: PhD position Causal Inference Solutions for the Prediction Paradox in Deployed AI-based Prediction Model
———
PhD position 3: Incorporating external data in clinical trials.
Help us enhance our understanding of issues related to the use of external data (e.g. from earlier trials, real world evidence) in randomized clinical trials. The project involves identifying practical and methodological barriers to implementation and conducting innovative methods research to eliminate these obstacles.
Supervisory team: Joost van Rosmalen, Floor van Oudenhoven, Maarten van Smeden
More info & apply at: PhD position ‘Use of external data in clinical trials’
