Off-the-grid data-driven optimization of sampling schemes in MRI
This addresses the need for improved sampling schemes in MRI, offering incremental advancements in versatility and constraint handling for medical imaging applications.
The paper tackles the problem of generating efficient and physically plausible sampling patterns in MRI by proposing a learning-based algorithm that works off-the-grid and handles arbitrary constraints, resulting in more versatile patterns that utilize all scanner degrees of freedom.
We propose a novel learning based algorithm to generate efficient and physically plausible sampling patterns in MRI. This method has a few advantages compared to recent learning based approaches: i) it works off-the-grid and ii) allows to handle arbitrary physical constraints. These two features allow for much more versatility in the sampling patterns that can take advantage of all the degrees of freedom offered by an MRI scanner. The method consists in a high dimensional optimization of a cost function defined implicitly by an algorithm. We propose various numerical tools to address this numerical challenge.