Meeting Banner
Abstract #3754

Compressive Manifold Learning Respiratory Self-Gated Liver MRI: Estimating the Respiratory Motion Directly from Undersampled K-Space

Muhammad Usman1, Paul Aljabar1, Tobias Schaeffter1, Claudia Prieto1, 2

1King's College London, London, United Kingdom; 2Pontificia Universidad Catolica de Chile, Santiago, Chile

Manifold learning approaches can be applied in MRI to extract meaningful patterns of variation from the high-dimensional set of images. An example is respiratory self-gating where the low dimensional respiratory signal can be extracted from a set of free breathing images. In this work, we propose Compressive Manifold learning for respiratory self-gated MRI. This approach estimates the respiratory signal directly from undersampled k-space data, without the need for image reconstruction. Results from simulated free-breathing liver MR data show that the respiratory signal can be accurately extracted from highly undersampled k-space samples using the proposed method. Free-breathing acquisitions, performed prospectively in 3 volunteers, also demonstrate the feasibility of CML respiratory self-gating in k-space.

Abstract PDF Presentation Video

Keywords

abdomen abdominal acceleration accuracy accurately achieved acquisition acquisitions addition allows applicable applications apply assuming beam black blue blurring breath breathing close clustering college combines combining comparable compressed compressive computed concept congress considering continuously corr correlated correlation cost cross curve curves diaphragm diaphragmatic dimension dimensional directly distances done embedded embedding ensuring estimating estimation even example extracting fact feasibility fidelity five fold frame frames frameworks free fully function gated gating gold golden green guarantees healthy highly hold holds hunter implies in vivo introduced just king learned learning linear liver manifold manifolds math matrix media minimized minimizing motion navigator need onto orange original partial partitioned pencil performance potentially preservation preserving press processing projected property propose proposed prospective pseudo radial random randomly real reconstruction reconstructions recover recovered remain removed requires resides respiratory restricted retrospective sampled sampling satisfies scaling scanner scheme schemes self sensing simulated simulations since space sparse spectral steps structure theory trajectories underlying volunteer volunteers yield