Role Overview
As a Computer Vision Intern, you will contribute to Nuclivision’s research and development activities focused on deep learning for PET and PET/CT imaging. We are offering two internship topics for motivated students with a background in AI, biomedical engineering, or computer science. Each intern will be assigned one of the following projects:
- Topic 1 – Data Augmentation in PET Imaging
Investigate and evaluate (generative) augmentation techniques for PET image enhancement and classification tasks. This includes reviewing and implementing augmentation strategies and analyzing their effects on model robustness and performance. - Topic 2 – Contrastive Learning for PET/CT
Research and apply contrastive learning methods to improve representation learning in PET/CT imaging. The aim is to support clinically relevant diagnostic applications and enhance model performance in downstream tasks such as denoising or classification.
Key tasks include:
- Conducting a literature review of relevant methods in medical image analysis.
- Implementing deep learning pipelines using Python and PyTorch.
- Training and evaluating models on real-world PET or PET/CT datasets.
- Reporting results in a structured and reproducible manner.
- Collaborating with Nuclivision’s AI and clinical teams.
Required Qualifications
- Bachelor’s degree (or ongoing studies) in computer science engineering, physics engineering, artificial intelligence, biomedical engineering, or a related field.
- Interest in medical imaging and deep learning, especially PET or PET/CT.
- Proficiency in Python and experience with deep learning libraries such as PyTorch or TensorFlow.
- Strong analytical and problem-solving skills.
- Full proficiency in English.
- Ability to work independently in a research-oriented setting.
What We Offer
- Work from the Wintercircus campus in Ghent at least one day per week.
- A dynamic startup environment with real-world exposure to AI in medical imaging.
- Access to clinical datasets and supervision by experienced AI researchers.
- The opportunity to contribute to ongoing product development or publications.
- A young, driven, and collaborative team to work with.
Location: Ghent, Belgium
Duration: 4 to 8 weeks
How to Apply
Please complete the form below and indicate in your motivation letter if you intend to undertake this internship as part of your student curriculum (for ECTS credits).