PhD Position: Computational Methods for Ultra-High Resolution Brain PET Imaging

You will join the Nuclear Medicine Research Group at KU Leuven / University Hospitals Leuven, an internationally renowned center in the field of medical imaging. The nuclear medicine / molecular imaging group has a long-standing track record in advanced molecular imaging, Positron emission tomography (PET) tracer development, kinetic modeling, and especially PET image reconstruction. The very multidisciplinary team consisting of physicists, engineers, computer scientists, chemists, and medical doctors works in close collaboration with global academic and industrial leaders and benefits from access to state-of-the-art infrastructure such as the NeuroExplorer PET/CT system, one of the most advanced brain PET scanners, with currently only three installations worldwide.
Website unit

Project

The position is part of a large-scale research project focused on exploiting ultra-high resolution PET to improve diagnosis and understanding of diseases of the central nervous system. The PhD student will contribute to:

• Development of event-by-event list-mode PET image reconstruction methods , including depth-of-interaction (DOI) modeling and motion correction techniques.
• Implementation of CT-free attenuation correction, using time-of-flight-based joint estimation methods and deep learning approaches.

The work will combine computational modeling, implementation and validation of advanced algorithms, and analysis of clinical and phantom data, with the aim of pushing the resolution and accuracy of PET imaging beyond current standards.

Profile

We are looking for a candidate with:

• a Master’s degree in Applied Mathematics, Physics, Engineering, Computer Science, or a related “technical” field.
• a solid foundation in calculus, linear algebra, and numerical methods.
• demonstrated problem-solving abilities and strong analytical thinking.
• scientific programming experience in at least one high level language such as Python, R, matlab, …
• the ability to “debug” complex problems
• a high degree of perseverance and the ability to work independently on challenging problem
• background knowledge in image reconstruction, inverse problems, or PET imaging is a plus, but not required. A bright and motivated candidate can be trained on the application side.

Offer

We offer:

• a fully funded 4-year PhD position at KU Leuven, ranked as one of Europe’s top universities
• daily collaboration with a multidisciplinary team of physicists, engineers, physicians, and computational scientists
• access to high-performance computing infrastructure, including state-of-the-art GPU servers
• the opportunity to work with one of the world’s first ultra-high resolution PET systems (NeuroExplorer)
• International networking opportunities through active collaborations with leading academic centers and industry partners
• competitive salary and benefits package according to KU Leuven standards.