Project Reference: TRG27-RAD-JK1

Project title: Integrating Structural MRI and Hyperpolarised Xenon MRI for Advanced Pulmonary Assessment

Project outline

Standard medical scans, such as CT and conventional MRI, produce detailed images of the human body but struggle to effectively image the lungs. Because the lungs are mostly filled with air, their magnetic resonance signals decay too rapidly for standard MRI receivers to detect. This project addresses this physical limitation by integrating two specialised imaging methods: Zero Echo Time (ZTE) MRI and Hyperpolarised Xenon MRI (XeMRI).

The primary objective of this PhD is to develop the acquisition and reconstruction methodologies required to unify these two techniques. ZTE MRI is an ultra-fast scanning method capable of capturing the rapidly decaying signals of hydrogen atoms in lung tissue, providing essential structural information. XeMRI involves the patient inhaling a safe, magnetised xenon gas, which produces strong signals used to accurately map lung function and ventilation. Developing the pulse sequences to acquire both data types will allow us to build a comprehensive model of pulmonary physiology.

A significant challenge of ZTE MRI is that its rapid, non-standard data sampling patterns often lead to image artefacts and missing data. To overcome this, the methodology will focus on designing and evaluating machine learning reconstruction pipelines. The project will train advanced AI tools, inspired by neural operators and end-to-end models like AUTOMAP, to accurately reconstruct missing data and generate clear, high-fidelity images.

Finally, the project will fuse the AI-reconstructed structural ZTE images with the functional ventilation maps generated by XeMRI. By directly correlating structural tissue abnormalities with functional gas distribution, this research aims to establish robust, quantitative biomarkers. This will provide clinicians and researchers with a precise, unified imaging tool for assessing and monitoring respiratory conditions.

Principal Supervisor

Dr Joshua Kaggie

Email address: jk636@cam.ac.uk

Department / Partner Institute: Radiology

Website: https://www.medschl.cam.ac.uk/joshua-kaggie

BBSRC strategic theme

Transformative technologies