My background is in mathematics and computer science. My expertise is in computational modelling, pattern recognition, and machine learning for biomedical imaging and data analysis.  My research aims to add understanding and capability in biomedicine by drawing on ideas from medical imaging, computer vision, data science, and machine learning. Much of my work focusses on imaging and the fusion of information from imaging with other data types (imaging + X).

The Microstructure Imaging Group works towards non-invasive histology. The idea is to use non-invasive imaging techniques, such as MRI to estimate features of tissue microstructure, such as cell size, shape and packing density, that traditionally require invasive biopsy and microscopy. Key applications are in brain connectivity mapping, white matter diseases like multiple sclerosis, and tumour-grading and treatment-planning for cancer. 

The POND group aims to learn the progression pattern of diseases - the appearance and development of symptoms and pathologies - from patient or population databases. Much of the work has focussed on neurodegenerative diseases, such as Alzheimer's disease and other dementias, using image data bases.

My work on image quality transfer (IQT) aims to estimate high quality from low quality images using machine learning. For example, we might estimate from a patient image acquired on a standard hospital scanner, the image we would obtain from the same patient in a high powered bespoke experimental scanner. The ideas help enable next-generation low-power portable imaging devices and enhance imaging capabilities in lower-and-middle income countries.