Daniel Alexander

Daniel Alexander

Professor of Imaging Science, Centre for Medical Image Computing, Department of Computer Science

University College London, United Kindgom

Title: Image quality transfer and democratization of MRI

Abstract: Daniel will talk about some international efforts to democratize MRI expertise and capability. For example the CAMERA network aiming to enhance MR education and research in Africa. Daniel will focus particularly on my work on Image Quality Transfer (Alexander et al Neuroimage 2017; Tanno et al Neuroimage 2021; Lin et al ISMRM 2021), which contributes to this effort. The technique learns models that infer high quality images, e.g. that we would have acquired from a one-off super-powered scanner, from lower quality images, e.g. acquired on a standard hospital scanner or a low-cost low-field scanner situated e.g. in an LMIC clinic. Initially designed for exploiting the rich information from one-off bespoke scanners such as the Connectom scanners (Jones et al Neuroimage 2018), the technique adapts naturally to enhance images from low-power scanners, such as low-field open-magnet or portable MRI scanners, to approximate images from standard high-field scanners. Daniel will talk through the history of development of these ideas, show some of the latest results, speculate about future opportunities, and describe some challenges and observations of implementing these ideas in LMIC scenarios (see link#1; and link#2).

Biography: Daniel Alexander is Professor in the Computer Science Department at UCL and Director of the UCL Centre for Medical Image Computing. His expertise is in computational modelling, machine learning and pattern recognition, mostly with application to medical data and in particular medical imaging data. He is best known for his work in diffusion MRI microstructure imaging techniques such as NODDI (Zhang et al Neuroimage 2012) for neuroimaging and VERDICT (Panagiotaki et al Cancer Research 2013) for cancer imaging, his work in disease progression modelling using techniques such as the event-based model (Fonteijn et al Neuroimage 2012; Young et al Brain 2014) and the subtype and stage algorithm (Young et al Nature Comms 2018), and more recently his work on image quality transfer (Alexander et al Neuroimage 2017; Tanno et al Neuroimage 2021; Lin et al ISMRM 2021). He has a BA in Mathematics from the University of Oxford (1993), an MSc in Computer Science from UCL (1994), and a PhD in Computer Science from UCL (1998). He worked as a post-doc at the University of Pennsylvania until 2000 when he returned to London to take up an academic position. He became full professor in 2009, Director of CMIC in 2015, and senior fellow of the ISMRM in 2017.