PhD Candidate in Explainable AI (xAI) for Capsule Endoscopy Norwegian University of Science and Technology in Norway

PhD opportunity at the Norwegian University of Science and Technology (NTNU) in explainable AI for capsule endoscopy.

This position is based in Gjøvik, Norway, in the Department of Computer Science at NTNU. The selected candidate will join the Marie Skłodowska-Curie Actions (MSCA) Doctoral Network CAPTURE, an international and intersectoral research project focused on advancing wireless capsule endoscopy through computer vision and artificial intelligence.

The research project addresses an important clinical challenge: current AI systems for pathology detection in capsule endoscopy can behave like black boxes. The PhD candidate will develop explainable AI (xAI) methods for automated diagnostic decision systems, with the goal of making predictions more understandable, actionable, and useful for medical doctors. The candidate will also work on evaluating explanations together with clinicians to assess relevance, usefulness, and clinical importance.

CAPTURE combines academia, hospitals, and industry partners across Europe. The network focuses on inflammatory bowel disease (IBD) and related gastrointestinal diseases, while supporting broader advances in wireless endoscopy. As a doctoral candidate in the network, you will participate in workshops, conferences, transferable skills training, and research secondments. Planned secondments include stays at Robert Gordon University, UK and Technical University of Darmstadt, Germany.

The position is full-time and fixed-term for 36 months, with an expected start date of 1 August 2027. The salary is normally NOK 580,000 per year, which covers the MSCA living and mobility allowances; a family allowance may also apply under MSCA rules. The position is conditional on external funding.

Applicants should have experience in computer vision and machine learning, strong Python programming skills, and familiarity with deep learning tools such as PyTorch or TensorFlow. A relevant master's degree in computer science, electrical engineering, or an equivalent field is required, together with a major independent project or thesis component. Strong academic performance and excellent English communication skills are required, and applicants must also satisfy the MSCA mobility rule.

Preferred background includes medical image processing, self-supervised learning, low-data learning, pretrained models, transfer learning, robustness analysis, uncertainty estimation, calibration, out-of-distribution detection, adversarial robustness, causal analysis, counterfactual explanations, or uncertainty quantification. Experience with scientific writing, open-source contributions, or a strong master's thesis is considered an advantage.

Applications must be submitted electronically via Jobbnorge.no by 19 November 2026. Required documents include transcripts and diplomas, CV, a copy or draft of the master's thesis, a project outline, a short motivation letter, relevant publications or certificates, and contact information for at least two referees. If education was completed outside Norway, additional documentation of the scope and quality of the degree(s) is required.

For questions about the position, contact Professor Kiran Raja at [email protected]. For recruitment questions, contact [email protected].