Engineering & Physics
Magnetic Resonance Elastography (MRE) is a non-invasive technique measuring the stiffness of soft tissues using MRI. Currently this technique is used for imaging the liver and cancer. We are developing new methodologies use MRE to assess cardiac/vascular and neuro‐psychiatric conditions. We are also extending MRE to develop Magnetic Resonance Force Imaging, a novel non‐invasive methodology for measuring cancer therapy response and the potential for tumours to metastasise.
Parallel Transmit (PTx) allows the improved image quality for MRI. Currently implanted metal objects (e.g. replacement joints and pacemakers) prevent MRI being conducted. We are exploring a safety concept with PTx called active decoupling, which would prevent energy absorption by these implanted metallic objects. This would also transform the availability of devices for MRI‐guided minimally‐invasive procedures such as cardiac arrhythmia ablation.
Generalised motion correction framework addresses the challenges of motion during the imaging process, e.g. the movement of the chest as we breathe or the movement of the heart beating. We are developing frameworks to estimate the motion from the image itself or from its data and also frameworks to create reconstructed images that have been corrected from any errors caused by motion.
Quantitative MR (qMR) gives information on the signal intensities of a scan for cardiac, cancer and brain tissue characterisation, as well as a visual representative of the area of the body being imaged. To enable this high-resolution 3D tissue mapping we are using recent advances to reduce the scan time of existing techniques and combining qMRI with the generalised motion correction framework outlined above. We are also using advances in MRI hardware, such as gradients capable of ultra-fast switching, to improve the contrast encoding strength and acquisition speed of these quantitative techniques.
Multi-modal imaging combining PET and MRI is possible using scanners that combine the two imaging modalities. We are using data from each modality to exploit the similarities between the data (such as anatomical structure) whilst preserving modality-unique information. These methods minimise the limitations of each individual modality and can generate both faster MRI acquisitions and higher quality PET images. This results in enhanced imaging and improved measurement of parameters such as metabolism and perfusion. This multi-modal imaging will also improve important clinical predictors such as treatment response in cancer.
Imaging robotics develops robotic systems that guide ultrasound in foetal and cardiac diagnostic imaging, and image-guided interventions. We are combining robotic-guided image acquisition with robotic-guidance of interventional devices for treatment of cardiac arrhythmias. We are also exploring application of artificial intelligence to enable automated image and catheter manipulation for carrying out treatments, reducing risk of complications, procedure time and radiation dose.