PhD physics-informed deep learning foundation models for cardiology

In this PhD project, you will develop physics-informed and multimodal deep learning foundation models for cardiology. The aim is to create innovative and explainable approaches that improve care for cardiology patients as part of a leading European consortium of medical centres and computer science groups. 

You will combine medical imaging, clinical data, and domain knowledge from physics-based models. Your work contributes to more accurate diagnoses, improved prediction of treatment outcomes, and personalized care for patients with cardiovascular diseases. 

Would you like to know more about the different phases within the PhD trajectory? You can read more about this on this page.

You will work at the intersection of artificial intelligence, medical physics, and clinical cardiology. You develop advanced deep learning models and translate them into clinical applications.

Your tasks include:

• developing and optimizing neural network models in Python (PyTorch);
• integrating domain knowledge from medical physics and cardiology into models;
• designing and testing innovative deep learning architectures;
• analyzing multimodal clinical data, including medical imaging;
• collaborating with researchers, clinicians, and AI specialists to implement solutions.

You are a curious and motivated researcher with an interest in both technical development and clinical application. You enjoy working in multidisciplinary teams and tackling complex challenges in healthcare AI.

You bring:

• an MSc degree in artificial intelligence, computer science, (medical) physics, mathematics, medical informatics, or a related field;
• solid object-oriented programming experience in Python and familiarity with building neural networks in PyTorch or similar frameworks;
• experience in designing, implementing, and experimenting with custom deep learning models beyond standard architectures;
• strong communication, presentation, and scientific writing skills in English;
• a strong interest in applying data science and AI to real clinical problems, preferably in cardiovascular imaging;
• the ability to quickly learn new technical concepts and collaborate effectively in multidisciplinary teams, including clinicians and physicists.

Experience with ECG, MRI or medical image data is a plus (but candidates with general data science/model development experience are strongly encouraged to apply).