XDMD

Master’s Thesis Intern – AI for Rare Cancers

Location: Utrecht
Employment type: full time
Deadline: 1 June 2026

One in five cancer patients in the Netherlands has a rare form of cancer – 20,000 people per year. Pathology plays a central role in diagnosis and treatment: determining whether cancer is present and how aggressive it is. Today, the analysis of pathology images is largely manual, and there is a growing shortage of pathologists worldwide. AI has the potential to make this faster, more accurate, and more robust.

In this project, you will work on an AI foundation model for pathology image analysis in rare cancers. A foundation model is a model trained on a large and diverse dataset that can be fine-tuned for a range of downstream tasks. Comparable foundation models have recently been developed for pathology images; none currently targets rare cancers specifically. The central methodological challenge is data scarcity: can a foundation model pre-trained on large general pathology datasets be effectively fine-tuned with few examples, and does it generalize across rare cancer types?

The concrete image analysis task will be defined together with a pathologist at the start of the project. Your work will span data collection and organization, model development, and validation across datasets and institutions.

About XDMD

XDMD is an engineering company in medical imaging. We build custom-made AI solutions in close collaboration with clinical and industrial partners for radiology, pathology, or any other domain where medical images are used. Our work combines strong engineering discipline with deep domain understanding and we advance innovations from concept through production.

What you will be working on

  • Review of existing foundation models for pathology (open-source and commercial)
  • Review of data-efficient fine-tuning methods for low-data regime, including few-shot and self-supervised learning approaches
  • Collection and organization of pathology image datasets from public, and where possible, clinical sources
  • Training and fine-tuning of a foundation model
  • Validation of the model across datasets and cancer types

Your profile

  • MSc student in AI, machine learning, computer science, biomedical engineering, or closely related field, enrolled at a Dutch university
  • Strong programming skills in Python
  • Interest in AI in medical imaging

Practical

  • 6 months paid graduation thesis project
  • Supervised by Dr.ir. Rashindra Manniesing (XDMD) and a pathologist
  • Based at our office at UtrechtInc, the startup incubator of Utrecht University
  • Applications are reviewed on a rolling basis
  • Potential start date: as soon as possible

Please send your CV and a brief motivation to vacancy@xdmd.ai, with “rare” as the subject line.