MSc Thesis: Vision Foundation Models for Medical Imaging

While foundation models have revolutionized natural language, computer vision—and medical imaging in particular—presents a fascinating set of open challenges. In this master thesis, you will train and benchmark vision foundation models to help advance radiotherapy in cancer care.

Project Description

Foundation models are trained on vast amounts of unlabeled data to learn broad, transferable representations. In this project, you will explore how self-supervised feature learning can be tailored to the complexities of medical imaging. You will train custom model variants on internal data and benchmark them against existing public models. The core focus is evaluating how effectively these unsupervised features transfer to key downstream tasks, such as image segmentation and generation.

Your main tasks

Review relevant research on vision foundation models and self-supervised learning in medical imaging.Set up, run, and manage multiple training iterations of custom model variants using internal and public datasets.Systematically benchmark the trained models against existing vision foundation models.Evaluate feature transferability to downstream tasks, such as image segmentation and generation.

We are looking for a curious master's student who wants to work where AI meets cancer care. You enjoy combining programming with careful experiments, and you look critically at results instead of taking them at face value. You are independent and well organized, and you can explain your findings clearly to both software engineers and medical physicists.

Ongoing master's studies in engineering physics, medical physics, computer science, applied mathematics or a related fieldCoursework or project experience in optimization and/or machine learningGreat programming skills in PythonHands-on experience with LLMsFluent written and spoken English

Application

Please apply to the position through the link below. Selection and interviews will be ongoing. We do not accept applications by e-mail.

Application Link.