PhD candidate 'Clinical and technical validation of Generative AI-based Digital Twins for cardiovascular imaging'

• Application Deadline: May 21 2026
• Nijmegen
• Hybrid

Job Description

Job description

Cardiovascular disease remains a leading cause of mortality while the predictive value of routinely acquired imaging and clinical data is still underutilized. Within the Horizon Europe TWIN-X project generative AI is used to build multimodal digital patient twins that integrate imaging clinical data and longitudinal health trajectories.

In this PhD you will focus on a critical and currently underdeveloped aspect: rigorous clinical validation of these systems. Rather than developing models your work will determine whether digital twins provide reliable generalizable and clinically meaningful predictions across institutions and patient populations.

You will design and conduct validation studies including multi-site external evaluations and clinical reader studies. Using large-scale datasets (coronary CT aortic CT cardiac MRI) you will assess model performance calibration robustness and behaviour under real-world parallel you will contribute to the development of structured reporting approaches that make AI outputs interpretable and reproducible in clinical workflows. You and your work are expected to be highly visible so youll be able to build presentation and reseach communication skills.

This position sits within a large international consortium of 19 partners across Europe including TU Munich University of Crete 3R Swiss Imaging Network EuSoMII and others. You will collaborate with leading groups in medical AI and contribute to widely used benchmarking infrastructures such as Grand Challenge.

Part of TWIN-X Horizon Europe Call: HORIZON-HLTHTOOL-03 19 partner institutions across 9 European countries

Preferred start date: 1 September 2026

Place of work

You will be based in the Department of Medical Imaging at Radboud University Medical Center (Radboudumc) embedded in the Diagnostic Image Analysis Group (DIAG). DIAG is one of Europes leading research groups in medical image analysis with researchers across Radiology and Nuclear Medicine Pathology and Cardiology. The group has a strong track record of publishing in high-impact journals. The working atmosphere is collaborative international and impact driven.

This project represents a novel dimension for the group: a bridge toward rigorous real-world clinical evaluation of AI. Radboudumc operates the Grand Challenge platform a globally used infrastructure for benchmarking medical imaging AI which you will work with directly.

You will be (hybridly but personally) supervised by Dr M. Huisman (PI) attending cardiovascular and musculoskeletal radiologist and Vice President Elect of EuSoMII with expertise in clinical AI validation post-market surveillance and standardization of AI as a regulated clinical technology. Your promotor is Prof. Henkjan Huisman (DIAG Radiology). You will work relatively independently embedded within a group with strong technical expertise in imaging AI and have regular interaction with international partners across the TWIN-X consortium. Hybrid working is an option depending on individual factors and project requirements.

At the department of Medical Imaging you work on diagnostics interventions education and scientific research. Our focus spans three key areas: Radiology Nuclear Medicine and Anatomy. Together with your colleagues you contribute to the early detection of diseases and the improvement of treatments helping us make healthcare more precise and patient friendly.