Learning a sampling pattern for MRIMS5

Taking measurements in MRI is a time-consuming procedure, so ideally one would take few samples and still recover a useable image. We consider the problem of learning a suitable sampling pattern for a class of objects that are similar. Given a training set consisting of pairs of clean images and MRI measurements, a bilevel optimisation problem can be formulated and studied, whose solution is a sparse sampling pattern that gives rise to good reconstructions.

This presentation is part of Minisymposium “MS5 - Learning and adaptive approaches in image processing (2 parts)
organized by: Kostas Papafitsoros (Weierstrass Institute Berlin) , Michael Hintermüller (Humboldt University and Weierstrass Institute Berlin) .

Ferdia Sherry (University of Cambridge)
image reconstruction, inverse problems, machine learning