Using quantitative magnetic resonance imaging (MRI), we are developing new mathematical and computational techniques that can identify early heterogeneous response of advanced cancers to novel therapies. Using these techniques we hope to be able accurately target regions that require additional treatments, including radiotherapy, in order to improve patient survival.

Using novel computational approaches, we are trying to better understand the biological mechanisms that lead to observed measurements from quantitative imaging. This includes diffusion-weighted MRI, which is thought to be inversely related to tumour cellularity and thus provides a potent indication of tumour cell kill following successful treatment.

We are exploiting new approaches in artificial intelligence (AI) and deep-learning to improve the patient pathway and accuracy of imaging studies within cancer. This includes (i) acceleration of image acquisition times, (ii) development of new approaches for predicting the occurrence of radiotherapy toxicities, and (iii) understanding how AI can be best used to understand quantitative imaging changes in response to treatment.