skip to content

Researcher: Angelica I. Aviles-Rivero, Noémie Debroux, Veronica Corona, M. Graves, G. Williams, C. Le Guyader, Carola-Bibiane Schönlieb

The conceptual philosophy of joint models (aka multi-tasking) for image analysis has demonstrated powerful results especially for complex data including medical. This improved performance is due to the fact that - by sharing representation between tasks and carefully intertwining them, one can create synergies across challenging problems and reduce error propagation. This results in boosting the accuracy of the outcomes whilst achieving better generalisation capabilities than sequential models and keeping reasonable computational cost. These advantages have motivated a fast development of new algorithmic approaches including hybrid techniques (variational + machine learning techniques/ model-based + data-driven methods).  We seek to show how multi-task methods offer better meaningful clinical outputs that traditional sequential approaches.

Selected Related Papers:

A.I Aviles-Rivero, G. Williams, M. Graves and C.B Schönlieb. CS+M: A Simultaneous Reconstruction and Motion Estimation Approach for Improving Undersampled MRI Reconstruction. ISMRM 2018.

Related Publications 

Compressed sensing plus motion (CS + M): A new perspective for improving undersampled MR image reconstruction
AI Aviles-Rivero, N Debroux, G Williams, MJ Graves, C-B Schönlieb – Medical image analysis (2021) abs/1810.10828, 101933
Multi-tasking to Correct: Motion-Compensated MRI via Joint Reconstruction and Registration
V Corona, AI Aviles-Rivero, N Debroux, M Graves, C Le Guyader, CB Schönlieb, G Williams – SCALE SPACE AND VARIATIONAL METHODS IN COMPUTER VISION, SSVM 2019 (2019) 11603 LNCS, 263