Danny Alexander

Title: Image quality transfer for low-field MRI and other imaging/AI applications

Abstract: I will talk through some recent applications and advances of machine learning and computational modelling for medical imaging at UCL Computer Science and UCL Hawkes Institute (formerly CMIC).  I will focus on on-going work on image quality transfer and its role in developing next-generation MRI platforms particularly for low-resource settings.  Image quality transfer aims to estimate a high-quality image, e.g. from a relatively lengthy acquisition or expensive scanner, given only a low quality image from a rapid or low-power device.  The ideas are important in the realisation of clinical applications for low-field MRI scanners, but challenges arise with modelling transformations from high field to low field and avoiding or mitigating hallucinations; see for example (Alexander et al Neuroimage 2017; Lin et al MedIA 2021; Kim et al ECCV 2024).

Bio: Danny Alexander is Professor in Imaging Science and head of Computer Science at UCL (since 2024). Previously, Prof Alexander was Director of the Centre for Medical Image Computing (CMIC: https://www.ucl.ac.uk/medical-image-computing/ucl-centre-medical-image-computing-cmic) 2015-2024. He initiated the Microstructure Imaging Group http://mig.cs.ucl.ac.uk and the Progression of Neurodegenerative Diseases initiative https://ucl-pond.github.io. Amongst other roles, Prof Alexander is also the theme lead for the UCLH Biomedical Research Centre Healthcare Engineering and Imaging theme https://www.uclhospitals.brc.nihr.ac.uk/our-research/healthcare-engineering-and-imaging-hei and holds a Wellcome Trust Investigator Award in Science.

Talk delivered as part of Sussex AI seminar series on 10th September 2025.