Abstract

Machine Learning has shown great potential in improving the entire medical imaging workflow, from image acquisition and reconstruction, to disease diagnosis and treatment. This talk will focus on the recent advances of deep learning in medical imaging, particularly on the inverse problems in medical imaging with applications to image reconstruction and image registration. Specifically, the talk will first describe how model-based deep learning can be used for MRI reconstruction from undersampled data and will discuss about our recent progress on deep learning-based reconstruction for accelerated dynamic cardiac MRI cine imaging. Then the talk will also show how deep learning can be leveraged to address the ill-posed inverse problem of image registration, where it will introduce our latest work in multi-/mono-modal image registration and myocardial motion tracking.

Join Zoom Meeting https://maths-cam-ac-uk.zoom.us/j/98746672780 Meeting ID: 987 4667 2780